JP4694424B2 - Authentication device - Google Patents

Description

  The present invention relates to an authentication device that authenticates an image by detecting a feature image such as a shape or color of a person, face, fingerprint, character, or the like as a subject, and in particular, exposes exposure values from the over side to the under side with the same composition. The present invention relates to an authentication apparatus including an imaging unit such as a camera having a bracket shooting function of shooting a plurality of still images by changing the direction or the opposite direction.

  As described above, the authentication device needs to detect a characteristic image such as the shape or color of a person, face, fingerprint, character, etc. as a subject as clearly as possible. A function of automatically controlling the exposure is provided so that the feature image can be detected as clearly as possible.

  In the automatic exposure (AE: auto exposure, automatic exposure) control in a general imaging device (camera) as well as the imaging device for the authentication device, for example, according to the state of the subject (and the operation input of the photographer) By analyzing the luminance signal value, an appropriate exposure value (EV: exposure value, exposure amount) is obtained. At that time, the imaging screen is divided into a plurality of photometric windows having a predetermined size, and a luminance signal value for each photometric window is selected and analyzed. For example, in the center-weighted metering AE control, the exposure value is determined using the luminance integrated value at the center of the imaging screen.

  Aperture priority AE is when the photographer first sets the aperture with the operation input and the camera controls the shutter speed. Conversely, the camera controls the aperture with the photographer first setting the shutter speed with the operation input. Is referred to as shutter priority AE. The film sensitivity (ISO sensitivity) of the camera is the ability to sense the light on the light receiving surface. If the ISO sensitivity is high, the shutter speed can be increased with aperture priority AE, and the aperture value can be narrowed with shutter priority AE. . Increasing the shutter speed is effective, for example, for camera shake countermeasures or shooting fast subjects.To narrow the aperture value, for example, increase the depth of field and focus from the near side to the far side. It is effective when you want.

  In the automatic exposure control of the camera, for example, when the brightness is 1000 lux and the appropriate exposure amount on the film surface (light receiving surface) is 1 lux · second giving a film density of 1.0, a gray plate with a reflectance of 18% The shutter speed and aperture value are controlled so that the amount of exposure is 1 lux.sec. If the brightness is 500 lux, the shutter speed and aperture value are controlled so that the exposure amount of a gray plate with a reflectance of 18% is 1 lux · second (the same exposure amount as 1000 lux) in that environment.

  The appropriate exposure value determined by the automatic exposure control by the camera is determined as described above. On the other hand, the appropriate exposure value of the image for the photographer, that is, “the right brightness of the image” for the individual photographer. May have different values depending on the sensitivity of each photographer and the composition of the subject. For example, “I want to make it brighter (overexposed)”, “I want to make it darker (underexposed)”, “I want to display a bright background without overexposure even if the subject in the foreground disappears darkly”, etc. The appropriate exposure value varies depending on the shooting intention. Furthermore, the sensitivity of the photographer and the composition of the subject may vary depending on complex shooting conditions, such as various imaging subjects, complex lighting environments such as daytime, nighttime, and indoors, and the dynamic range of the imaging device is narrow compared to the lighting conditions in the surrounding environment. Therefore, the appropriate exposure value of the image changes. Therefore, there is a limit to obtaining an appropriate exposure value for the photographer by the AE control function of the camera.

  In particular, when shooting under complicated shooting conditions or when high-precision exposure control is required, an appropriate exposure value for the photographer may not be obtained with the AE control function (including fixed exposure) set in the camera. In such a case, it is difficult to obtain an appropriate exposure value by manual operation because the photographer needs experience and skill.

  Bracketing (bracketing) is a technology that was originally used with a steel camera using a silver salt film. In the case described above, that is, when it is difficult to determine the appropriate exposure manually during shooting, When proper exposure cannot be obtained with the fixed exposure or automatic exposure settings, first, in the same composition, set the exposure value in descending order from the over side (eg +2 EV) to the under side (eg -2 EV) or vice versa. This is a method of taking a plurality of still images by changing (switching) the direction in a stepwise manner (for example, every 2 EV) with a uniform width. The function that the camera automatically performs bracket shooting by setting the number of exposure steps to change is called auto bracketing, and the photographer can select the optimal exposure from the multiple images of the same composition. By selecting this image, it is possible to finally obtain an image with proper exposure.

  Some imaging devices with an AE control function have an automatic exposure (AEB: auto exposure bracketing) function that captures images while automatically switching the camera exposure with the number of frames set for bracket shooting as described above. Is provided. In recent digital cameras (electronic still cameras), products with an AEB function have been announced, but in general, digital cameras perform image processing on each captured image before recording it in memory each time one is taken. Since time is required, bracketing with a digital camera often requires a longer processing time than a silver halide film camera.

  As a conventional electronic still camera having such a bracket shooting function, for example, an operation unit for inputting a bracket shooting instruction by a photographer, and an exposure change circuit for correcting exposure of shot image data, Image data with different exposure values can be obtained by correcting the initial image data taken with the camera setting value or the photographer setting value in a stepwise uniform manner according to the exposure correction level specified for bracket shooting. An imaging device to generate is known (for example, see Patent Document 1).

  When an instruction for bracket shooting is input to the operation unit of the imaging apparatus of Patent Document 1, the exposure change circuit according to the exposure correction level specified in bracket shooting for the image data shot according to the instruction input. Image data with corrected exposure is generated. Therefore, it is possible to obtain a plurality of still images with the exposure value changed stepwise with the same composition without actually shooting a plurality of images, and to shorten the shooting time of bracket shooting.

  In addition, as an electronic still camera having another conventional bracket shooting function, there is an imaging apparatus that obtains image data having different ISO sensitivity values by changing the ISO sensitivity value instead of the exposure EV value and automatically bracketing the image data. It is known (see, for example, Patent Document 2). This imaging apparatus includes an analog gain adjusting means for adjusting gain at an analog signal stage, an exposure level adjusting means for performing level correction and gamma correction at a digital signal stage, and a correction level of an ISO sensitivity value designated for bracket shooting. And a control means for correcting by combining both adjusting means so as to be uniform in steps.

  The brightness of the captured image changes depending on the amount of exposure (exposure amount), and the brightness of the captured image also changes depending on the ISO sensitivity value. In some cases, the ISO sensitivity value may be controlled instead of the above control. Therefore, bracket photography is possible even if the ISO sensitivity value is controlled as described above.

  Moreover, since the S / N ratio of the analog gain adjusting means generally deteriorates as the sensitivity becomes higher (high gain), noise may become conspicuous. In addition, the exposure level adjustment means can make noise on the high sensitivity side unnoticeable by noise reduction image processing, but if the sensitivity adjustment (digital gain adjustment) is carried out significantly, the gradation in the highlight area or shadow area will be lost. In some cases, the dynamic range decreases, resulting in an unnatural image. Therefore, in the imaging apparatus of Patent Document 2, by controlling the ratios of the adjustment amounts of the analog gain adjustment unit and the exposure level adjustment unit according to the shooting conditions and the correction direction, the S / N ratio, the gradation characteristics, and The dynamic range was improved.

Japanese Patent Laid-Open No. 11-4380 (page 6-8, FIG. 1) JP 2005-142607 A (pages 21-23, FIG. 4)

  In the conventional imaging apparatus, the exposure correction value in the image data obtained by shooting a plurality of images by bracket shooting is, for example, exposure with respect to the initial exposure value obtained by the automatic exposure function of the camera or the manual operation of the photographer. A correction value in which the value is changed with a fixed width of one or more steps at equal intervals, such as +/- 1 EV or +/- 0.5 EV, or +/- equal intervals specified by the photographer The correction value is changed with a fixed width. This is a correction value that is changed with a fixed width at +/− equal intervals similarly designated by the photographer even in the case of correction by ISO sensitivity values as in Patent Document 2.

  In that case, for example, even when the optimal correction value is in the middle of two correction values changed by a fixed width, the correction is made so as to match either of the two correction values, and the intermediate value of the two correction values Could not be corrected. That is, there is a problem that the accuracy of the exposure correction value cannot be increased.

  In particular, when shooting a subject in a dark place, image data with overexposure corrected by bracket shooting will be selected, but the correction range is uniform at regular intervals. If the accuracy of the is not high, if the exposure is slightly under, it was easy to overcorrect the value at once. In that case, when the analog gain adjustment becomes a high gain, the S / N ratio deteriorates, and there is a problem that the image deteriorates.

  In order to increase the accuracy of correction in the conventional imaging apparatus, for example, the correction width of the equal interval (fixed value) for changing the exposure is narrowed to 1/2 or the like, and the number of shots is doubled accordingly. For example, a method of increasing the number can be considered. However, in this case, there is a problem that extra imaging processing time is required corresponding to the increase in the number of shots, and the time required for bracket shooting becomes long. Since the digital camera requires time for image processing after shooting, there is a problem that the time for bracket shooting is particularly long.

  The above-described problem of the imaging device that performs bracket photography becomes particularly significant when the imaging device is applied to, for example, an image authentication device. For example, in an authentication device that detects a feature image such as a shape or color of a person, face, fingerprint, or character as a subject and authenticates the match with the registered content, the accuracy of the exposure correction value cannot be increased. That is, the recognition accuracy of image details cannot be increased, and the authentication accuracy cannot be improved.

  In addition, the image authentication apparatus often shoots a subject in a dark place. Therefore, an excessively high gain often increases the S / N ratio and deteriorates the authentication performance in many cases. There is a problem of becoming.

Furthermore, the image authentication apparatus needs to increase the image recognition accuracy in order to increase the recognition accuracy. In this case, for example, as described above, the correction for changing the exposure of the imaging device at a fixed value at equal intervals. It is conceivable to reduce the width. However, in that case, the number of shots increases. In addition to the need for imaging processing time corresponding to the increased number of shots, extra authentication processing time is also required. The authentication processing time is a process for detecting, for example, in which position in which image data the subject to be authenticated is imaged, or in which image data the subject being imaged is an appropriate exposure. That is, in an image authentication apparatus using an imaging apparatus that performs bracket shooting, there is a problem that the recognition processing time becomes long if the recognition accuracy is improved.
As described above, it is very difficult to set an exposure correction value so that an authentication result is optimal in an image authentication device using an imaging device that performs bracket shooting.

  The present invention has been made in order to solve the above-described problems, and is an authentication device using an imaging device that performs bracket shooting, which improves authentication accuracy and reduces authentication performance due to deterioration in image quality. An object of the present invention is to provide an image authentication apparatus in which an increase in a series of recognition processing time from imaging to authentication is suppressed.

An authentication apparatus of the present invention includes a recording unit that stores in advance image data of subjects used for authentication of a plurality of subjects,
Imaging means for imaging a subject by bracket shooting for shooting a plurality of images with normal shooting or changing exposure, and outputting an image signal of the captured image;
Feature image detection availability determination means for detecting a predetermined feature image in the captured image from the image signal of the captured image and outputting at least a feature image detection availability signal capable of determining whether the feature image has been detected;
A feature image exposure correction value for correcting exposure by detecting a predetermined exposure correction signal in a predetermined feature image region including the feature image of the captured image is generated from the image signal and the feature image detection enable / disable signal. Exposure correction value detection means for
A first exposure control value for the first shooting of normal shooting or bracket shooting is calculated from the integrated value of the photometric window of the predetermined dimension, and bracket shooting is calculated from the first exposure control value and the characteristic image exposure correction value. A second exposure control value for the second and subsequent shootings with different exposures is calculated, and the exposure at the time of shooting is controlled based on the first exposure control value or the second exposure control value. Exposure control means for outputting an exposure control signal;
An authentication unit that determines whether or not authentication is possible using a captured image that has been shot with exposure controlled by the exposure control unit and image data for authentication stored in the recording unit ;
Equipped with a,
The authentication means includes
When the feature image is not included in the photometric window of the predetermined size, exposure is controlled using the exposure control signal based on the first exposure control value and the second exposure control value. Using the captured image and the image data for authentication stored in the recording means to determine whether authentication is possible,
In the case where the feature image is included in the photometric window having the predetermined size, a captured image obtained by controlling exposure using the exposure control signal based on the first exposure control value, and the recording unit To determine whether authentication is possible using the authentication image data stored in
It is characterized by that .

  According to the authentication apparatus of the present invention, a feature image such as a shape or color is detected using a person, face, fingerprint, character, or the like as a subject, and a signal indicating the exposure state of the feature image is used for exposure control of bracket shooting. Therefore, it is possible to provide an image recognition apparatus that improves the authentication accuracy, has little degradation in authentication performance due to image quality degradation, and suppresses an increase in a series of recognition processing times from imaging to authentication.

  The bracket shooting described in each of the following embodiments will be described as a case in which three frames are automatically bracketed while automatically changing the exposure according to the set exposure correction amount.

Embodiment 1 FIG.
FIG. 1 is a block diagram showing a schematic configuration of the authentication apparatus according to the first embodiment of the present invention.
The authentication apparatus of FIG. 1 basically includes an imaging unit 51 that captures an image to authenticate a subject, a recording unit 9, a feature image detection availability determination unit 14, an exposure correction value detection unit 15, an authentication unit 52, and a system control unit 24. Configured as Detailed functions of each part will be described later.
1 includes an image selection unit 42, a feature image area detection unit 25, an average luminance detection unit 26, and an exposure correction value generation unit 41. The feature image of the luminance signal from the digital signal processing unit 5 is provided. Information on the captured image corrected by inputting the correction output for detection VF is input, and the feature image detection enable / disable output VG from the feature image detection enable / disable determining unit 14 is input, thereby correcting the captured image. A feature image region including a feature image is detected from the image, and a feature image exposure correction value VH is output from an average luminance signal value as an exposure correction signal for the feature image region (feature image).

  The lens 1 is a means that constitutes the authentication apparatus of FIG. 1 and condenses incident light from a subject (not shown) on the light receiving surface of the solid-state imaging device 2. The solid-state image pickup device 2 photoelectrically converts incident light from the subject and outputs it as an image pickup output VA that is an electrical signal of RAW output depending on the pixel arrangement of the solid-state image pickup device 2. The analog signal processing means 3 performs correlated double sampling processing (CDS) and automatic signal amplification processing (AGC) on the imaging output VA, and outputs the amplified output VB. The A / D conversion means 4 converts the amplified output VB, which is an analog signal, into a digital signal in units of image frames, and outputs it as an A / D conversion output VC.

The digital signal processing means 5 has a function of performing image processing on the imaging output (A / D conversion output VC) that has become a digital signal. Hereinafter, the digital signal processing means 5 will be described in detail with reference to FIG.
FIG. 2 is a block diagram showing an internal configuration of the digital signal processing means 5 of FIG.
The integrating means 30 calculates an integrated value obtained by integrating the A / D conversion output VC for each photometric window having a predetermined size formed by dividing one screen in the horizontal direction and the vertical direction, and calculates the integrated value to the CPU bus. Memory control is also performed so that the data is output to the SDRAM 11 via. The photometric window having a predetermined size is determined in advance by design, or is set by the user through a user interface. The integrated value includes a RAW signal RAW integrated value corresponding to the pixel array of the solid-state imaging device 2, a luminance integrated value of a luminance signal obtained from the RAW signal, and the like.

The pixel interpolation unit 31 interpolates pixels from the input A / D conversion output VC using an A / D conversion clock, and generates an RGB format video signal from the interpolated signal.
The YCbCr conversion processing unit 32 converts the RGB video signal into a YCbCr video signal that can be processed by the digital signal processing unit 5.
The Y signal processing means 33 performs signal processing on the luminance signal such as enhancement processing of the luminance signal (Y signal) in the video signal in the YCbCr format.

  The gradation conversion means 34 corrects the gradation of the Y signal (luminance signal) and outputs the corrected Y signal to the feature image detection availability determination means 14 and the external interface means 35 as a feature image detection correction output VF. As described above, in the present embodiment, the Y signal (luminance signal) of the captured image is output as the feature image detection correction output VF of the digital signal processing means 5, and the feature image detection correction output VF is also described below. Is a luminance signal. The feature image detection correction output VF may be any signal that can be used to detect a feature image by the feature image detection availability determination unit 14 in addition to the luminance signal of the present embodiment. A color difference signal (CbCr), a primary color RGB signal, or the like may be output.

  The external interface unit 35 is a display output signal VI (for example, YCbCr4: 4: 2 output format, RGB output format) that can process a video signal in YCbCr format by an external device such as the display unit 8 or the recording unit 9 connected in the subsequent stage. ) And output. In addition, the external interface unit 35 may be provided with an image quality correction function for performing image quality correction so as to improve the display performance of the display device when connecting to an external device.

  The chroma signal processing means 36 performs signal processing for adjusting the hue and saturation of a chroma signal (CbCr signal) such as a 2 × 2 matrix in the video signal in the YCbCr format.

The timing signal generating unit 6 is a synchronization control unit that synchronizes the solid-state imaging device 2, the analog signal processing unit 3, the A / D conversion unit 4, and the digital signal processing unit 5. Further, by extending the memory function for the digital signal processing means 5, a configuration of asynchronous control is possible.
For example, the timing signal generating means 6 generates the driving timing pulses for horizontal driving and vertical driving via the V driver 7 for the solid-state imaging device 2 as shown by the timing signal TS1 in FIG. As shown in the timing signal TS2, the drive timing pulses for charge reading (TG) and charge sweeping (SUB) are generated and output.

  Further, the timing signal generation means 6 also generates an analog signal processing pulse for performing the CDS processing as shown in the timing signal TS3 of FIG. 1 for the analog signal processing means 3, and generates the timing signal TS4 of FIG. As shown, an A / D conversion clock is generated for the A / D conversion means 4, a drive pulse is generated for the digital signal processing means 5, and the aperture is reduced as shown in the timing signal TS5 of FIG. A drive pulse is generated for 13.

  More specifically, for example, when the first exposure timing control signal ES (1) for normal photographing is input to the timing signal generating unit 6, the timing signal generating unit 6 uses the first normal photographing photographing signal. The timing signal TS1 (1) and the timing signal TS2 (1) are output to the solid-state imaging device 2, the first timing signal TS3 (1) for normal shooting is output to the analog signal processing means 3, and the first signal for normal shooting is output. One timing signal TS4 (1) is output to the A / D conversion means 4, and the first timing signal TS5 (1) for normal photographing is output to the aperture 13. When the second exposure timing control signal ES (2) for bracket photographing is input to the timing signal generating means 6, the timing signal generating means 6 uses the second timing signal TS1 (2) for bracket photographing. And the timing signal TS2 (2) is output to the solid-state imaging device 2, the second timing signal TS3 (2) for bracket photographing is output to the analog signal processing means 3, and the second timing signal TS4 ( 2) is output to the A / D conversion means 4, and the second timing signal TS5 (2) for bracket photographing is output to the aperture 13.

  The recording unit 9 holds a video signal of a captured image captured by the individual imaging device 2 and also stores verification image data for verification with the captured image by an authentication unit 52 described later.

  The exposure control means 10 reads the integrated value of the A / D conversion output VC for each photometric window held in the SDRAM 11 via the CPU bus to detect the exposure state. In normal photographing, the exposure control means 10 corresponds to the above-mentioned photometric window. Implement automatic exposure control. During bracket photography, exposure is controlled as will be described in detail below using FIG.

FIG. 3 is a block diagram showing an internal configuration of the exposure control means 10 of FIG.
The first exposure control value calculation means 38 reads and inputs the integrated value obtained by integrating the image data signal proportional to the output signal of the solid-state imaging device 2 in the SDRAM 11 via the CPU bus. The first exposure control value is calculated, and the first exposure control value is output to the exposure control signal generation means 40 and the second exposure control value calculation means 39. The first exposure control value is a value for controlling the exposure for the normal shooting or the reference shooting for the bracket shooting. In the normal shooting, the first exposure control value is used for controlling a value (exposure value) converged in the automatic exposure control. Is the value of The second exposure control value calculation means 39 receives the characteristic image exposure correction value VH from the exposure correction value detection means 15 and the first exposure control value from the first exposure control value calculation means 38 and receives the second exposure control value calculation means 39. The exposure control value is calculated, and the second exposure control value is output to the exposure control signal generation means 40. In the exposure control signal generation means 40, the first exposure control value is input from the first exposure control value calculation means 38 or the second exposure control value is input from the second exposure control value calculation means 39, and exposure timing control is performed. A signal ES is generated and output to the timing signal generator 6, and an exposure gain control signal EG is generated and output to the analog signal processor 3.

  The timing signal generating means 6 controls the exposure by outputting the timing signal TS2 so as to control the number of electric charge sweep pulses (SUB pulses) of the electronic shutter according to the received exposure timing control signal ES, and performs analog signal processing. A timing signal TS3 for controlling the timing of the means 3 is output. On the other hand, the analog signal processing means 3 adjusts the gain at the timing of the timing signal TS3 in accordance with the received exposure gain control signal EG.

  The SDRAM 11 is connected to the CPU bus and temporarily stores data from the control device and image files. The flash memory 12 is also connected to the CPU bus and stores software for the control device.

  The feature image detection availability determination unit 14 uses a feature image detection correction output VF of the luminance signal of the input captured image, for example, using a pattern matching technique, for example, a predetermined feature image in the image, specifically, for example, a person, Detect feature images such as faces, eyes, shapes, and characters. As the feature image detection availability output VG, a signal capable of determining whether or not at least a feature image has been detected is output. If possible, a signal indicating the level of clarity, reliability, accuracy, etc. of the detected feature image Output the value.

In the above description, pattern matching is used as the feature image detection method of the present embodiment. However, the feature image detection method applicable to the present invention is not limited to this, and a feature image is input from an input captured image. Any other detection method can be used as long as it can be detected.
For example, a skin color detection technique can be used to detect the feature image. In this case, the luminance signal and the color difference signal are input from the digital signal processing unit 5 to the feature image detection availability determination unit 14 as the feature image detection correction output VF. Then, the feature image detection possibility determination means 14 detects a feature image from the feature image detection correction output VF of the luminance signal and the color difference signal, or at least a signal that can be determined whether or not the feature image has been detected, or is detected. A feature image detection availability output VG which is a signal value indicating the level of clarity, reliability, accuracy, etc. of the feature image is output.

  The feature image detection availability determination unit 14 receives the captured image captured by the bracket from the imaging unit 41, and extracts a captured image from which an optimum feature image can be detected from each captured image captured by the bracket. Based on the extracted ID (recognition code) of the captured image, the exposure control means 10 generates an exposure control signal corresponding to the captured image from which an optimum feature image can be detected. A captured image subjected to exposure control using an exposure control signal optimal for detection of the feature image is output to the subsequent stage. By using a captured image that is optimal for the detection of the feature image, the subsequent processing time can be shortened. When the latter stage is authentication processing, the authentication time of the captured image can be shortened.

FIG. 4 is a block diagram showing an internal configuration of the exposure correction value detecting means 15 of FIG.
Inside the exposure correction value detection means 15, an image selection means 42, a feature image area detection means 25, an average luminance detection means 26 and an exposure correction value generation means 41 are provided, and the characteristics of the luminance signal from the digital signal processing means 5 are provided. The information of the captured image corrected by inputting the correction output for image detection VF is input, and the feature image detection enable / disable output VG from the feature image detection enable / disable determining unit 14 is input to be corrected. A feature image region including a feature image is detected from the captured image, and a feature image exposure correction value VH is output from an average luminance signal value as an exposure correction signal for the feature image region (feature image).

The image selection unit 42 receives the feature image detection correction output VF and the feature image detection enable / disable output VG of each luminance signal of a plurality of captured images whose exposure values are changed by bracket imaging, and corrects the feature image detection of the luminance signal. From the output VF, the feature image detection correction output VF of the brightness signal of the picked-up image optimum for detecting the feature image exposure correction value VH is selected using the feature image detection result by the feature image detection enable / disable output VG. It is output to the feature image area detection means 25.
The feature image area detection unit 25 receives the feature image detection correction output VF of the luminance signal of the captured image selected by the image selection unit 42, and the feature image is detected from the feature image detection correction output VF of the luminance signal of the captured image. A feature image area including (person, face, eye, shape, character, etc.) is detected.
Further, for example, in the case of the feature image area detection means 25 in the case of the above-described skin color detection, a signal value area that can detect that the luminance signal value and the color difference signal value are skin colors is detected as the feature image area. In this way, a feature image area including a feature image can be detected from the feature image detection correction output VF of the brightness signal of the captured image or the brightness signal value and the color difference signal value. Is output.
The average luminance detecting means 26 receives the characteristic image detection correction output VF of the luminance signal and the signal indicating the characteristic image area including the characteristic image from the characteristic image area detecting means 25, and the exposure correction signal in the characteristic image area. As a result, an average luminance signal value is detected.

  The exposure correction value generation unit 41 calculates a feature image exposure correction value VH used for correcting the exposure of the feature image in the feature image area from the average brightness in the feature image area detected by the average brightness detection unit 26. Generated and output to the exposure control means 10.

FIG. 5 is a diagram showing a case where a single captured image including an example of the feature image area according to the first embodiment of the present invention is displayed on the display unit 8.
The feature image area in the image of the present embodiment will be described below with reference to FIG.

In the image 16 in which one screen imaged by the authentication apparatus of FIG. 5 is displayed on the display means 8, a human image of a subject is arranged near the center of a horizontally long rectangular screen, and the peripheral portion of the human image is Is included and displayed. As the feature image area 17 of FIG. 5, a substantially rectangular area including the face of the person image of the subject and a part of the surrounding area is extracted.
The feature image region 17 is a region including the image of the feature image 18 in the image 16, and is a region used for performing exposure correction described later. Then, the feature image 18 of the present embodiment is an eye in a person image of a subject as shown in FIG. The feature image area 17 of the present embodiment is not limited to a substantially rectangular area as long as it includes the feature image 18 in the face, and may be a circular area or the like. Further, an area narrower than the area shown in FIG. 5 may be extracted. In this case, that is, when the feature image region 17 is changed to only a narrow region including the feature image 18, the accuracy of the average luminance signal value in the region used for correcting the exposure can be increased.

  The system control unit 24 is, for example, a control unit such as a microprocessor or a CPU, and controls the entire system in the authentication device. For example, power ON / OFF control, drive mode switching control, display control on the display device 8 is performed. Etc. Further, the system control unit 24 sets the drive mode of the imaging unit 51 of the normal imaging drive and the bracket imaging drive according to the feature image detection period by the feature image detection availability determination unit 14, the authentication period by the authentication unit 52, and the normal drive time. Switch.

  1 will be described in detail below with reference to the timing chart of FIG. 6 and the flowcharts of FIGS. 7 to 12.

FIG. 6 is a timing chart showing the operation timing at the time of bracket shooting in the authentication apparatus of the present embodiment.
FIG. 6A shows the vertical synchronization signal VD of the individual image sensor 2 in the authentication device. Between each vertical synchronizing signal VD is a transfer period of one frame of image data in the solid-state image sensor 2, and FIG. 6A shows a transfer period of frames F0 to F7F8.
FIG. 6B shows a charge readout pulse TG of the individual image sensor 2. During the transfer period for each frame (hereinafter referred to as a frame transfer period), the charge is read once at the charge read timing T0 by the charge read pulse TG. The read charges are transferred in the charge transfer processing period T1, and after the transfer is completed, an exposure control signal ES / EG is obtained in the exposure detection processing period T2, and exposure is controlled during the subsequent frame transfer period. The

  FIG. 6C shows the charge discharge pulse SUB. Charges are accumulated in the individual image sensor during the period from the end of the charge discharge pulse SUB to the next charge read timing T0. The exposure time can be adjusted by controlling this period. In this embodiment, the exposure control of the shutter priority AE of the electronic shutter is described. However, the relationship between the charge readout timing (T0), the charge transfer processing period (T1), and the exposure detection processing period (T2). The same applies to the exposure control of the shutter priority AE of the mechanical shutter, for example.

  FIG. 6D shows a bracket shooting request REQB. A situation where bracket shooting is necessary is detected by the system control device 24 of FIG. 1, and a bracket shooting request REQB is output from the system control device 24. The situation where bracket shooting is necessary is, for example, when there is a bracket shooting instruction input by the user's operation, and the bracket shooting is performed by the imaging device (imaging means: camera) from the luminance integrated value and exposure value for each photometric window. Is automatically specified, when there is a request for bracket shooting from a host server (not shown), or when the imaging device is used for an authentication device, the counter or timer is operated periodically (for example, For example, when there is a task request for authentication bracket shooting performed every second).

FIG. 6E is a diagram showing an exposure control value for each frame by the A / D conversion output VC, which is an image signal converted into a digital signal for each frame.
The exposure control value for frame: F0 in FIG. 6 (a) is shown by EP (F0) in FIG. 6 (e), and the exposure control value EP (F3) for frame: F3 is sequentially from the exposure control means 10. The image data of the exposure state according to the first exposure control value is output as the A / D conversion output VC with a delay of one frame period. That is, the first exposure control value obtained by the exposure control means 10 from the image data up to the i-th frame (i is an integer of i ≧ 0) is the exposure control value of the A / D conversion output VC of the i + 1th frame. EP (F (i)) is output.

  In FIG. 6 (e), the exposure control value of frame: F0 is delayed by EP (F0) by one frame period, the exposure control value of frame: F1 is delayed by one frame period, EP (F1), and the exposure control value of frame: F2 Is output by EP (F2) with a delay of one frame period, and the exposure control value of frame: F3 is output by EP (F3) with a delay of one frame period. Up to this point, the image data of the exposure state based on the first exposure control value is obtained. Thereafter, the exposure control value for frame: F4 is delayed by EP (F3) × α1 after a delay of one frame period, the exposure control value for frame: F5 is delayed by EP (F3) × α2, and the exposure control value of frame: F6. Is delayed by one frame period, and the exposure control value of frame F7 is output by EP (F6) * α4 delayed by one frame period. These are the image data of the exposure state by the second exposure control value.

  When the bracket shooting request REQB is input at the timing of FIG. 6D, the A / D conversion output VC of the image data of the exposure control value EP (F3) next to the exposure control value EP (F2) at that time is At the same time as it is output, it shifts from normal shooting to bracket shooting. When the bracket photographing is performed, the first exposure control value calculation unit 38 of the exposure control unit 10 performs the first exposure control so as to be proportional to the input signal of the integration value integrated by the integration unit 30 of the digital signal processing unit 5. An exposure control value is calculated, and a plurality of second exposure control values having different exposure states are calculated by the second exposure control value calculation means 39 based on the first exposure control value, and the second exposure control value is calculated. The exposure control signal generation means 40 generates and outputs a bracket shooting timing and exposure gain control signal (second exposure control signal). Then, the system control device 24 frames the captured image of the exposure state reflecting the first exposure control value in the previous frame by the automatic exposure control and the captured image of the exposure state reflecting the second exposure control value. Switched in units.

In FIGS. 6A to 6F, the number of times of bracket imaging in the feature image detection period is set to 3 times, but the number of times of bracketing imaging can be set to 2 times or more.
The feature image detection result VK of the feature image detection availability determination unit 14 is input to the system control unit 24. If the feature image cannot be detected from the captured image, the probability that the feature image can be detected and the exposure image that can be recognized by the feature image can be obtained by increasing the number of times of bracket shooting with changed exposure. Increase.

  When changing the number of times of bracket shooting, the exposure control value set at the same time can also be changed. For example, when the number of brackets is 6 and the exposure control value correction width is 1 EV (exposure value), for example, the exposure control value can be corrected in 1/5 EV (0.2 EV) steps. . Here, exposure control correction values are assigned with uniform widths, but the steps of the correction width are changed as 0 EV, 0.25 EV, 0.4 EV, 0.5 EV, 0.6 EV, and 1.0 EV. May be.

  The processing mode in FIG. 6F shows a feature image detection period, an authentication period, and a normal imaging period. Here, the image capturing period is the automatic exposure shooting, and the feature image detection period and the authentication period indicate the case where the bracket shooting process is performed. The drive mode of the feature image detection period and the authentication period can be changed depending on the configuration of the authentication apparatus.

  FIG. 7 is a flowchart showing an overall processing operation of the authentication apparatus of FIG. Initial processing for initial setting of the authentication device (initializing the register in the imaging device and performing automatic control for several frames) is performed (S500), and processing for determining the processing mode of the authentication device in units of frames is performed. First, it is determined whether or not the imaging process has been performed (S501). If the imaging process has not been performed (S501: No), the imaging process (S503) is performed. When the imaging process is performed (S501: Yes), it is determined whether or not the feature image detection process is performed (S504). When the feature image detection process is not performed (S504: No), it is determined. A feature image detection process (S506) is performed. When the feature image detection process is being performed (S504: Yes), the authentication process (S508) is performed. Thus, the authentication apparatus has three processing modes.

  Further, in the previous stage of each processing mode, the system control unit 24 can switch the driving mode of the imaging apparatus when performing each processing mode. In FIG. 7, the drive mode of normal shooting (S502) is selected by confirming that there is no bracket shooting instruction (REQB) before the imaging processing (S503), and bracket shooting is performed before other processing modes. By confirming that there is an instruction (REQB), the drive mode of bracket shooting (S505, S507) is selected. In the case of the processing mode of the feature image detection process (S506) and the authentication process (S508), normal shooting may be selected as each drive mode.

  As described above, the feature image detection availability determination unit 14 detects feature images of a plurality of different captured images and outputs a feature image detection availability output VG according to the detection result. The exposure correction value detection means 15 selects a captured image suitable for exposure correction from a plurality of captured images using the feature image detection enable / disable output VG, and the feature image of each luminance signal in the feature image area of the selected captured image. The average brightness is detected from the detection correction output VF, and a feature image exposure correction value VH is generated from the average brightness of the feature image area and output to the second exposure control value calculation means 39. Therefore, more specifically, the second exposure control value calculation means 39 of the exposure control means 10 has a plurality of different exposure states for bracket photographing based on the characteristic image exposure correction value VH and the first exposure control value. A second exposure control value is calculated. Then, the exposure control signal generating means 40 of the exposure control means 10 uses a plurality of exposure timing control signals ES (2) for bracket photography and an exposure gain control signal EG (2) (second) based on the second exposure control value. Exposure control signal) and output to the timing signal generating means 6 and the analog signal processing means 3, respectively.

  That is, the first exposure control value is obtained by calculating the exposure control value from a signal proportional to the input signal of the integrated value integrated by the integrating means 30 of the digital signal processing means 5, whereas However, the difference is that the average brightness signal value of the region is detected using the feature image detection result, and the exposure control value is calculated so as to correct the first exposure control value.

  Note that the exposure control signal generating means 40 controls the exposure timing control for normal shooting based on the first exposure control value which is the exposure control value of the previous frame by the automatic exposure control when the bracket shooting request REQB is not input. The signal ES (1) and the exposure gain control signal EG (1) (first exposure control signal) are continuously generated for each frame. Thereby, the exposure value of the captured image is continuously switched.

  The authenticating unit 52 collates the feature image for matching captured by bracketing with the image data of the feature image for matching held in the recording unit 9 during the authentication period in the processing mode of FIG. It is determined whether it can be determined that the images are the same. If it can be determined that the image of the subject is the same as the image data for verification, information indicating that the subject has been authenticated is output. If the image is not determined to be the same, information indicating that the subject cannot be authenticated is output. The feature image detection method in the authentication means 52 is the above point in comparison with the operation (method) in which the feature image detection availability determination means 14 detects the feature image during the feature image detection period in the processing mode of FIG. Are different, the others are equivalent.

  The authentication unit 52 captures the feature image exposure correction value VH or the exposure control signal obtained by the feature image detection possibility determination unit 14, the exposure correction value detection unit 15, and the exposure control unit 10 of the preceding process of the imaging unit 51. Authentication is performed using the image. An image for performing authentication by collating with the captured image is temporarily held in the SDRAM 11. By using the feature image exposure correction value VH or the exposure control signal obtained by the feature image detection enable / disable determination unit 14, the exposure correction value detection unit 15 and the exposure control unit 10, the time until the authentication image is acquired is shortened. Can be made.

  The authentication unit 52 has a larger image size than the image for each photometric window of the feature region including the feature image used in the feature image detection possibility determination unit 14, the exposure correction value detection unit 15, and the exposure control unit 10 in the preceding process. By using the captured image, it is possible to authenticate other feature images, and to improve the authentication accuracy. Further, since the image size used in the feature image detection possibility determination unit 14, the exposure correction value detection unit 15 and the exposure control unit 10 is small in a part of the captured image, the transfer time of the image data from the image sensor can be shortened. The signal processing time of the image can be shortened. Accordingly, it is possible to shorten the time required for determining the exposure control signal necessary for performing authentication by the authentication apparatus.

  In the present embodiment, since it is described as a configuration in which three frames are taken as the bracket shooting period, in FIGS. 6A to 6E, the frames F4 to F6 indicate the bracket shooting request period. Note that the number of frames specified for bracket shooting is not limited to the three frames shown in the present embodiment, but may be any frame of two or more.

FIG. 8 is a flowchart showing a processing operation in the imaging apparatus of the authentication apparatus of FIG.
FIG. 9 is a flowchart showing a normal photographing process based on the first exposure control value / first exposure control signal in the authentication apparatus of FIG.
FIG. 10 is a flowchart showing a first exposure control value detection process in the authentication apparatus of FIG.
FIG. 11 is a flowchart showing bracket photographing processing based on the second exposure control value / second exposure control signal in the authentication apparatus of FIG.
FIG. 12 is a flowchart showing a first exposure control value detection process in the authentication apparatus of FIG.

In normal photographing, the exposure control means 10 automatically controls exposure according to a photometric window having a predetermined size designed in advance or a photometric window having a predetermined size set by a user through a user interface.
In FIG. 8, when the authentication apparatus is powered on, an initial process is started (S100).
In the initial processing, the register in the authentication device is initialized and automatic control processing for several frames is performed, so that an initial first exposure control signal (exposure timing control signal ES (1) and exposure gain control signal EG is obtained. (1)) is determined.
When the flowchart of the authentication device in FIG. 7 is executed, the initial exposure is performed by initializing the register in the imaging device in the initial processing (S500) in FIG. 7 and performing automatic control of several frames. Since the control value is determined, the initial process (S100) of FIG. 8 is not performed.

  After the initial processing (S100) in FIG. 8 is completed, the authentication apparatus detects the presence / absence of a bracket photographing request (normal photographing processing or bracket photographing processing) for each frame unit (S101). When there is no bracket shooting request (S101: No), the process proceeds to the normal shooting process (S102), and after executing the first exposure control value detection process (S103), the process returns to step S101 again to determine whether there is a bracket shooting request. Detection is performed.

  In the normal photographing process (S102), as shown in FIG. 9, a charge accumulation start process (S201), a charge accumulation end detection process (S202), and a charge transfer process (S203) are performed. The charge accumulation period from the charge accumulation start process (S201) to the charge accumulation end detection process (S202) is the period from the end of the SUB pulse in FIG. 6 (c) to the TG pulse (charge read) in FIG. 6 (b). The charge accumulation amount can be adjusted by controlling the charge accumulation period, that is, the exposure time. The charge transfer process (S203) is indicated by a charge transfer process period T1 in FIG.

  Next, in the first exposure control value detection process (S103) of FIG. 8, the integration process (S210) is performed by the integration means 30 of the digital signal processing means 5 as shown in FIG. The first exposure control value calculation means 38 performs the first exposure control value detection process (S211).

When the first exposure control value is obtained by the first exposure control value calculation means 38 of the exposure control means 10 in the integration process (S210) of FIG. 10, the photometry window is obtained by the integration means 30 of the digital signal processing means 5. An integrated value obtained by integrating the accumulated charge is detected every time.
At the same time, histogram detection of the maximum value, minimum value, and luminance signal value for each photometric window is performed by a part of the function of the digital signal processing means 5.

  In the first exposure control value detection process (S211) in FIG. 10, the exposure control means 10 obtains a first exposure control value from the accumulated value obtained in the accumulation process (S210), and then further performs the first exposure control. This is a process for obtaining signals (exposure timing control signal ES (1) and exposure gain control signal EG (1)). When the first exposure control value is set in the authentication device by the exposure timing control signal ES (1) and the exposure gain control signal EG (1), as shown in FIG. This is a case where control is performed.

Considering the case where the imaging device of the present embodiment is used for an authentication device, for example, when the feature image region 17 of FIG. 5 is included in the photometric window of the authentication device, the first exposure control signal is the feature image 18. Since there is a high possibility that it is an effective value for detecting, the need for bracket photography is low.
However, if the feature image area 17 is not included in the photometric window, there may be blackout or signal saturation in the feature image area 17, and the brightness sufficient to detect the feature image 18. There is a high possibility that the signal value or contrast is not satisfied. Therefore, in the authentication apparatus according to the present embodiment, when a feature image area is not included in the photometric window, bracket shooting is necessary in order to improve the recognition accuracy of the feature image. To implement. The feature image in this case is an image for setting a signal value for correcting exposure in an image based on the first exposure control signal for normal photographing.

  When the determination of whether or not the bracket shooting request in FIG. 8 is detected (S101) is performed (S101: Yes), bracket shooting processing (S106) is performed. When bracket shooting is first performed, exposure is corrected for bracket shooting using the first exposure control value. Thereafter, the bracket request is canceled (S107), and after the bracket photographing request is canceled, the second exposure control value detection process (S104) is performed.

  In the second exposure amount detection process (S104), an exposure correction value considering the exposure control value for normal shooting is obtained. After the second exposure control value detection process, the process returns to the determination (S101) as to whether or not a bracket shooting request has been detected again.

  In the bracket shooting process (S106), the first exposure control value obtained in the normal shooting and the first exposure control value are statistically optimized based on the experimental results considering the environment and the exposure conditions in advance. A second exposure control value for bracket shooting is obtained, and bracket shooting is performed using the second correction value.

  When the bracket photographing process (S106) ends, the bracket request is canceled (S107), and then the second exposure control value generation process (S104) is performed. As the second exposure control value detection process (S104), as shown in FIG. 12, a feature image detection availability determination process (S403), an image feature detection process (S405), and a second exposure control value generation process (S404). ) Is performed by the feature image detection possibility determination unit 14, the exposure correction value detection unit 15, and the exposure control unit 10.

  The exposure control signal reflecting the characteristic image exposure correction value VH set in the second exposure control value detection process (S104) is set as a second exposure control signal (FIG. 6E). The second exposure control signal is repeatedly generated every time a bracket shooting request is input, and is repeated until a captured image capable of detecting an optimum feature image is obtained. In bracket shooting of the present embodiment, by increasing the exposure control value to increase the number of bracket shooting, the method of obtaining a single captured image that is optimal for feature image discrimination (authentication), For example, a method of converging the exposure value by repeating a predetermined number of bracket shootings to detect a feature image and generate a feature image exposure correction value VH is possible. Note that when shooting is performed with an increased number of brackets, the exposure value can be adjusted more finely by setting a characteristic image exposure correction value VH that is obtained by dividing the adjustment range of the second exposure control signal according to the number of shots. Is possible.

  In FIG. 8, the second exposure control value detection process (S104) is after it is determined whether or not a bracket shooting request (REQB) has been input (S101). In the normal initial state, normal imaging is performed. Since only the first exposure control value detection process (S103) is performed first, the normal photographing process time is shortened.

FIG. 13 is a diagram showing a luminance signal waveform when the feature image region is detected by the exposure correction value detection means 15 and the feature image exposure correction value VH is obtained from the average luminance signal value.
In FIG. 13, the vertical axis represents the output signal value of the luminance of the image 16 in FIG. 5, and the horizontal axis represents the horizontal position on the image 16 in FIG. Yb1 to Yb2 are ranges of luminance signal values for optimum feature recognition statistically obtained by experiments for recognizing feature images from images, and Yb1 represents a luminance signal value at the lower boundary, Yb2 indicates the luminance signal value at the upper boundary. The average luminance signal value at each position of the broken line 19 in the feature image area 17 is shown. In FIG. 13, the broken line shows the luminance signal value Yav1 before correction by the exposure control of the normal imaging process, and the solid line shows the bracket imaging process. The luminance signal value Yav2 after the exposure is corrected by the exposure control is shown. A portion where the luminance signal value at the center of the luminance signal value Yav1 indicated by the broken line and the luminance signal value Yav2 indicated by the solid line is higher than that at the left and right ends corresponds to a face portion where the luminance is relatively high in the feature image area is doing.

First, after the power is turned on, as a result of performing exposure control in the normal photographing process, the characteristic image area average luminance signal value of the face portion in the characteristic image area 17 becomes Yav1. Α1 which is the underexposed feature image exposure correction value VH is obtained from the following expression (1) from the range of the luminance signal value for feature recognition described above, and is the overexposed feature image exposure correction value VH. α2 is obtained from the following equation (2).
α1 = K1 × Yb1 / Yav1 (1)
α2 = K2 × Yb2 / Yav1 (2)

  Here, the correction coefficients K1 and K2 in Expression (1) are numerical values obtained by experiments and statistical methods, K1 is a correction coefficient of 0 ≦ K1, and K2 in Expression (2) is K2 ≦ 1. Correction coefficient. Note that the feature image exposure correction value VH is not limited to one set for each of the above-described underexposure side and overexposure side, and has an arbitrary number of feature image exposure correction values VH according to the number of bracket shots. Also good.

  Subsequently, when a bracket shooting request is input, bracket shooting processing (S106) is executed. In the bracket shooting process (S106), as shown in FIG. 11, regarding the bracket shooting count, whether the bracket shooting count is initially set, whether the bracket shooting count itself is detected, or whether the bracket shooting count is compared with a specified value. The only difference is that the process of increasing the bracket shooting count by 1 (S310, S311, S316, S317) and the process of setting the second exposure control value according to the bracket shooting count (S312) are different. The processing (S313, S314, S315) is the same processing as the normal imaging processing (S201 to S203) shown in FIG. When the prescribed number of brackets have been photographed (S316), the bracket request is canceled (S107), and then the second exposure control value detection process (S104) is performed.

  FIG. 14 is a diagram for explaining the characteristic image exposure correction value VH of bracket shooting, and an example of the first normal imaging screen (shooting 1) and the corrected bracket imaging screen (shooting 2, 3) during bracket shooting. Is shown.

  In the imaging state 1, the average brightness level (APL) of the feature image area 17 by automatic exposure control similar to that of normal shooting and the APL of the photometric window whose exposure is controlled to be constant by being selected by the automatic exposure control of normal shooting. These show the state of each image of bracket shooting when they are almost the same. The imaging state 2 shows the state of each image of bracket shooting when the APL of the feature image area 17 is lower than the APL of the photometric window. The imaging state 3 shows the state of each image of bracket shooting when the APL of the feature image area 17 is higher than the APL of the photometric window.

Since the imaging state 1 in FIG. 14 is when the APL of the feature image area 17 and the APL of the photometric window are substantially the same, the signal of the feature image area 17 is also in an optimal exposure state. The values α1 and α2 of the characteristic image exposure correction value VH in such an imaging state are set so as to follow the following formula (3).
α1 ≦ 1 ≦ α2 (3)
(When K1 = K2 = 1)

  Then, on the bracket imaging screen in which the exposure value is corrected by the feature image exposure correction value α1 on the underexposure side, the feature image area 17 appears dark as shown in the shooting 2 in the imaging state 1. In addition, on the bracket imaging screen in which the exposure value is corrected by the overexposed feature image exposure correction value α2, the feature image area 17 appears brighter as shown in shooting 3 in the imaging state 1.

  By setting the feature image exposure correction value VH and performing bracket shooting in this manner, three imaging screens of a dark imaging result, a bright imaging result, and an intermediate imaging result are obtained. A feature image is detected from the three imaging screens. If, for example, a feature image having a luminance sufficient for image authentication is detected, the process ends. If a feature image with sufficient brightness for image authentication cannot be detected, change the exposure condition of the first captured image, or change the feature image exposure correction value, and then perform bracket shooting again, which is sufficient for image authentication. A feature image of brightness is detected.

The imaging state 2 in FIG. 14 is a case where the APL of the feature image region is lower than the surrounding APL, the contrast of the feature image is low, and there is a tendency to black out, so the value α1 of the feature image exposure correction value VH, By setting α2 to 1 or more according to the following formula (4), it is possible to suppress blackout and make the contrast clear.
1 ≦ α1 ≦ α2 (4)
(When K1 = K2 = 1)

The imaging state 3 in FIG. 14 is a case where the APL of the feature image area is higher than the surrounding APL, and the feature image or the periphery tends to be white and tends to be white and is difficult to authenticate. By setting the VH values α1 and α2 to 1 or less according to the following equation (5), it is possible to suppress whiteout and make the contrast clear.
α1 ≦ α2 ≦ 1 (5)
(When K1 = K2 = 1)

  Furthermore, even when a feature image can be detected by the first bracket photographing, the feature image exposure correction value VH in which the feature image can be detected (in the direction of increasing the correction value or in the direction of decreasing the correction value). By performing bracket shooting using different feature image exposure correction values VH, the detection accuracy can be further improved.

  In this way, the exposure state of the feature image area is detected from each imaging result of bracket shooting by APL, and exposure correction is performed according to the APL of the feature image area, thereby detecting the feature image and optimal exposure for image authentication. The optimum feature image exposure correction value VH can be obtained in a short time so that the detection accuracy of the feature image can be increased. Further, since the feature image can be detected with high accuracy, the authentication accuracy of the feature image can be improved during image authentication.

Embodiment 2. FIG.
The authentication device according to the present invention is not limited to the configuration of the authentication device according to the first embodiment shown in FIG. 1, and for example, realizes the same function even in a configuration in which a CPU bus is further effectively used as a signal and information transmission path. Can do.

FIG. 15 is a block diagram showing a schematic configuration of the authentication apparatus according to the second embodiment of the present invention.
In FIG. 15, when compared with the first embodiment shown in FIG. 1, the feature image detection availability determination unit 14 and the exposure correction value detection unit 15 are connected to the CPU bus, and signals and information are transmitted through the CPU bus. Is different. Other configurations are the same as those in the first embodiment.

  The digital signal processing means 5 outputs an integrated value and an output signal used for the feature image detection possibility determination means 14 (a signal equivalent to the feature image detection correction output VF of the first embodiment) to the CPU bus and is connected to the CPU bus. Held in the SDRAM 11.

  The feature image detection availability determination unit 14 is input with the captured image captured by the bracket from the imaging unit 41 via the CPU bus.

  In the SDRAM 11, the feature image detection availability output VG of the feature image detection availability determination unit 14 and the feature image exposure correction value VH output from the exposure correction value detection unit 15 are held in the SDRAM 11.

  The exposure control means 10 obtains a first exposure control signal (ES (1), EG (1)) for normal photographing using the integrated value held in the SDRAM 11, and uses the characteristic image exposure correction value VH. Then, a second exposure control signal (EG (2) / ES (2)) for bracket shooting is obtained to correct the exposure during bracket shooting.

  Compared with the authentication unit 52 of the first embodiment, the authentication unit 53 includes the feature image detection correction output VF in addition to the image data of the captured image and the feature image detection availability determination unit 14 and exposure correction via the CPU bus. The difference is that the value is input to the value detection means 15. Other operations are the same.

  By adopting such a configuration, the exposure control means 10, the exposure correction value detection means 15, the feature image detection availability determination means 14, and the authentication means 53 are controlled by a single microprocessor (for example, the system control means 24). Therefore, the board area of the authentication device can be reduced. In addition, by controlling the exposure control means 10, the exposure correction value detection means 15, and the feature image detection possibility determination means 14 by a microprocessor, advanced image processing can be realized, and the revision of the algorithm and the performance improvement can be facilitated. it can.

Embodiment 3 FIG.
The authentication apparatus of the present invention is configured to perform bracket shooting by controlling exposure with an open loop type using the output of the digital signal processing means 5 such as the authentication apparatus of Embodiments 1 and 2 shown in FIGS. However, the present invention is not limited to this case, and it is possible to use a closed loop type in which an image displayed on the display unit 8 is detected by an external sensor unit to control exposure and perform bracket shooting.

In the following, this embodiment in which bracketing photography is performed with exposure controlled by a closed loop type will be described with reference to the system configuration of FIG. 16 and the flowchart of FIG.
FIG. 16 is a block diagram showing a schematic configuration of the authentication apparatus according to the third embodiment of the present invention.
In the authentication apparatus according to the first embodiment illustrated in FIG. 1 and the authentication apparatus according to the present embodiment illustrated in FIG. 16, a subject to be authenticated within the imaging range of the authentication apparatus from the image displayed on the display device 8. The difference is that an external sensor means 23 for detecting whether or not there is added is added. The functions of the other constituent blocks to which the same reference numerals are assigned are the same.
In the case of the configuration of the second embodiment shown in FIG. 15, the same effect can be realized by adding the external sensor means 23 as in the present embodiment.

FIG. 17 is a flowchart showing an overall processing operation in the authentication apparatus of FIG.
In FIG. 17, in the initial process (S108), in addition to the process similar to the initial process (S100) of the first embodiment, the initial value of the feature image exposure correction value VH in the bracket control for exposure control is obtained from statistical processing by experiment. The obtained value is set.

  In the external sensor detection process (S109), it is detected from the image displayed on the display device 8 by using the external sensor means 23 whether or not there is a subject to be authenticated within the imaging range of the authentication device. If not detected (S109: No), the process of step S109 is repeated again. When the subject to be authenticated is detected by the external sensor means 23 (S109: Yes), the system control means 24 detects whether or not there is an exposure bracket photographing request (whether it has been output) (S110).

When the output of the bracket shooting request is detected (S110: Yes), the authentication apparatus performs shooting in the first exposure stage of the bracket shooting process (S111) as shown in step S106 of FIG. . If not detected (S110: No), the process of step S110 is repeated again.
After the image of the first exposure stage is taken, bracket photography is further performed using the initial feature image exposure correction value VH, and the feature image detection availability determination unit 14 uses the photographed images in the captured image. When the feature image detection enable / disable determination process (S112) for detecting whether the feature image exists is performed and the feature image can be detected (S112: Yes), the exposure correction value detection unit 15 reproduces the exposure correction value. A success detection process (S113) is performed.
In the system configuration like the configuration of the second embodiment shown in FIG. 15, the system control unit 24 can perform this function by using the CPU bus.

  After the feature image is detected by the feature image detection availability determination unit 14 (S112: Yes), it is detected that the exposure value of the feature image needs to be optimized (regeneration of the exposure correction value) (S113). : Yes), the feature image exposure correction value VH is generated again by the exposure correction value detection unit 15, and exposure control is performed by the exposure value control unit 10 and the timing signal generation unit 6. By regenerating the exposure correction value in this way, the feature image detection and authentication processing can be made highly accurate. In the present embodiment, the imaging process after regeneration of the feature image exposure correction value VH is the bracket imaging process, but may be a normal imaging process.

  If it is detected that the exposure value of the feature image is not optimized (regeneration of the exposure correction value) (S113: No), the feature image exposure correction value VH is regenerated by the exposure correction value detection means 15. It is not necessary to perform exposure control after regeneration, and feature image detection and authentication processing can be speeded up.

  Next, when the feature image cannot be detected by the feature image detection enable / disable determining unit 14 (S112: No), the process returns to step S110 to detect the output of the exposure bracket photographing request, and the initial feature image exposure correction value VH is regenerated. Is set. The bracket imaging process (S111) is performed again with the feature image exposure correction value VH after the resetting, and the feature image detection availability determination unit 14 determines whether the feature image can be detected from the bracket imaging process result (S112).

  In the present embodiment, the feature image detection availability determination process (S112) is performed after the bracket photographing process (S111), so that the accuracy of the feature image detection and the authentication process can be improved.

Embodiment 4 FIG.
FIG. 18 is a block diagram showing a schematic configuration of the authentication apparatus according to the fourth embodiment of the present invention.
In the authentication apparatus according to the first embodiment shown in FIG. 1 and the authentication apparatus according to the present embodiment shown in FIG. 18, the feature image detection availability determination means 14, the exposure correction value detection means 15, and the exposure control means 10 are realized. The difference is that a means for controlling the digital signal processing means 43 from the exposure control means 10 is added. The functions of the other constituent blocks to which the same reference numerals are assigned are the same.

FIG. 19 is a block diagram showing the internal configuration of the digital signal processing means 5 of FIG.
The gradation signal converting means 37 is different between the digital signal processing means 5 of the authentication apparatus of the first embodiment shown in FIG. 2 and the digital signal processing means 5 of the authentication apparatus of the present embodiment shown in FIG. In the gradation conversion unit 37 of FIG. 19, a plurality of gradation conversion characteristic tables are held in order to perform gradation conversion by inputting the output of the Y signal processing unit 33 and to output a feature image detection correction output VF. These gradation conversion characteristics tables are switched by the gradation control signal ET. The functions of the other constituent blocks to which the same reference numerals are assigned are the same.
The digital signal processing means 5 shown in FIG. 19 can be applied to the configuration of the second embodiment shown in FIG. 15 or the third embodiment shown in FIG. be able to.

FIG. 20 is a block diagram showing the internal configuration of the exposure correction value detection means 15 of FIG.
The exposure correction value detection means 15 of the authentication apparatus of the first embodiment shown in FIG. 4 and the exposure correction value detection means 15 of the authentication apparatus of the present embodiment shown in FIG. Yes. The functions of the other constituent blocks to which the same reference numerals are assigned are the same.

  The contrast detection means 29 detects and outputs the difference in contrast value in the feature image area shown in FIG. In accordance with the detected contrast difference, the exposure correction value generation means 41 generates a feature image exposure correction value VH. The numerical value of the contrast difference is a numerical value obtained by statistically processing the experimental data.

If the value of the contrast difference in the feature image region (for example, the difference between the luminance signal value of the feature image and the average luminance signal value in the feature image region) is smaller than the threshold for contrast detection, the current exposure value is It can be determined that the exposure state is difficult to recognize the feature image. In that case, since the captured image is highly likely to be blacked out, the values α1 and α2 of the characteristic image exposure correction value VH are set to 1 or more according to the following equation (6).
1 ≦ α1 ≦ α2 (6)
(When K1 = K2 = 1)

  When there is a high possibility that the captured image is blacked out, the gradation conversion means 37 in FIG. 19 uses the gradation control signal ET to change the gradation conversion characteristic (for example, the output signal becomes large when the input signal is small). , Γ = 2.0).

If the contrast difference in the feature image region is about the same or larger than the threshold for contrast detection, it can be determined that the exposure state is sufficient to recognize the feature image. In this case, since the captured image is likely to be a standard photographing result, the value α1 of the feature image exposure correction value VH is set to 1 or less according to the following equation (7), and the value α2 of the feature image exposure correction value VH. Is set to 1 or more according to the following equation (7).
α1 ≦ 1 ≦ α2 (7)
(When K1 = K2 = 1)

  In this case, when the captured image is a standard photographing result, the gradation conversion means 37 is operated by the gradation control signal ET when the input signal is large or the gradation conversion characteristic in which the input and output characteristics are linear. Is set so as to have a gradation conversion characteristic (for example, γ = 0.7) at which the output signal becomes small.

  Thus, by switching the feature image exposure correction value VH for bracket shooting according to the contrast difference of the feature image region, the feature image detection accuracy can be improved and the detection time can be shortened.

  If the contrast difference of the feature image area is small, the A / D conversion output VC may be saturated if the feature image exposure correction value VH is increased. Therefore, when there is a high possibility of saturation due to the feature image exposure correction value VH and APL, the tone conversion characteristics of the tone conversion means 37 are switched so that the feature image can be easily detected. By switching the gradation conversion characteristics, there is an effect that the feature image can be detected without saturating the A / D conversion output.

Embodiment 5 FIG.
In Embodiment 4 described above, the exposure is corrected using the contrast difference in the feature image area. However, in the initial state or in a state where the feature image is not recognized, the detection range of the contrast difference is expanded beyond the feature image area. In this embodiment described below, an average luminance signal value and a minimum luminance signal value or a maximum luminance signal value in an enlarged region of the feature image area are detected, and the detection result is determined in accordance with the detection result. A case where the characteristic image exposure correction value VH is controlled will be described.

For example, when the contrast difference between the minimum luminance signal value of the enlarged region and the average luminance signal value is small, the values α1 and α2 of the characteristic image exposure correction value VH are set to 1 or more according to the following equation (8).
1 ≦ α1 ≦ α2 (8)
(When K1 = K2 = 1)

When the contrast difference between the minimum luminance signal value of the enlarged area and the average luminance signal value is equal or large, the characteristic image exposure correction value VH is set to 1 or less according to the following formula (9), and the characteristic image exposure is performed. The value α2 of the correction value VH is set to 1 or more according to the following equation (9).
α1 ≦ 1 ≦ α2 (9)
(When K1 = K2 = 1)

  In the present embodiment, in addition to the same effects as in the fourth embodiment, the feature image detection accuracy can be improved and the detection time can be shortened even in the initial state or in a state where the feature image is not recognized. Have.

Embodiment 6 FIG.
FIG. 21 is a block diagram showing a schematic configuration of the authentication apparatus according to the sixth embodiment of the present invention.
The configuration of the present embodiment is different from FIG. 1 only in that the authentication unit 54 is connected via the network unit 55, and the other configurations are the same as those in FIG. The structure of is shown.

  The network means 55 is means for connecting the authentication device to the network.

  The authentication unit 54 is a unit that performs authentication via a network. This authentication means is configured as another hardware such as a personal computer or a special signal processing device.

  This embodiment also has the same effect as each of the above-described embodiments, and the degree of freedom such as the location of the authentication unit is increased through the network.

  In each of the above embodiments, the case of correcting the exposure of bracket shooting using the average luminance signal value and the contrast value has been described. However, the present invention is not limited to this, and for example, an exposure correction value detection unit. The S / N ratio is obtained by detecting “signal value” and “noise or dispersion” in 15 internal blocks, and the S / N ratio is used as a reference for detection to correct the exposure of bracket shooting. Also good.

It is a block diagram which shows schematic structure of the authentication apparatus of Embodiment 1 of this invention. It is a block diagram which shows the internal structure of the digital signal processing means 5 of FIG. It is a block diagram which shows the internal structure of the exposure control means 10 of FIG. It is a block diagram which shows the internal structure of the exposure correction value detection means 15 of FIG. FIG. 8 is a diagram showing a case where a single captured image including an example of a feature image area is displayed on the display unit 8 in the authentication apparatus of FIG. 1. It is a timing chart which shows the timing of the operation | movement at the time of bracket imaging | photography in the authentication apparatus of FIG. It is a flowchart which shows the whole processing operation in the authentication apparatus of FIG. It is a flowchart which shows the processing operation in the imaging device in the authentication apparatus of FIG. 7 is a flowchart showing a normal photographing process based on a first exposure control value / first exposure control signal in the authentication apparatus of FIG. 1. It is a flowchart which shows the detection process of the 1st exposure control value in the authentication apparatus of FIG. It is a flowchart which shows the bracket imaging | photography process based on the 2nd exposure control value / 2nd exposure control signal in the authentication apparatus of FIG. It is a flowchart which shows the detection process of the 1st exposure control value in the authentication apparatus of FIG. It is the figure which showed the signal waveform of the brightness | luminance when the characteristic image area | region is detected by the exposure correction value detection means 15 of FIG. 1, and the characteristic image exposure correction value VH is calculated | required from the average luminance signal value. It is a figure for demonstrating the characteristic image exposure correction value VH of bracket imaging | photography. It is a block diagram which shows schematic structure of the authentication apparatus of Embodiment 2 of this invention. It is a block diagram which shows schematic structure of the authentication apparatus of Embodiment 3 of this invention. It is a flowchart which shows the whole processing operation in the authentication apparatus of FIG. It is a block diagram which shows schematic structure of the authentication apparatus of Embodiment 4 of this invention. It is a block diagram which shows the internal structure of the digital signal processing means 5 of FIG. It is a block diagram which shows the internal structure of the exposure correction value detection means 15 of FIG. It is a block diagram which shows schematic structure of the authentication apparatus of Embodiment 6 of this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Lens, 2 Solid-state image sensor, 3 Analog signal processing means, 4 A / D conversion means, 5 Digital signal processing means, 6 Timing signal generation means, 7 V driver, 8 Display means, 9 Recording means, 10 Exposure control means, 11 SDRAM, 12 FLASH memory, 13 Aperture, 14 Feature image detection means, 15 Exposure correction value detection means, 16 Screen, 17 Feature image area, 18 Feature image, 19 Line indicating signal value used for explanation of FIG. 13, 23 External Sensor means, 24 system control means, 25 feature image area detection means, 26 average brightness detection means, 29 contrast detection means, 30 integration means, 31 pixel interpolation means, 32 YCbCr conversion processing means, 33 Y signal processing means, 34 gradation Conversion means, 35 external interface means, 36 chroma signal processing means, 3 First exposure control value calculation means, 39 Second exposure control value calculation means, 40 Exposure control signal generation means, 41 Exposure correction value generation means, 42 Image selection means, 51 Imaging means, 52, 53, 54 Authentication means, 55, network means, EG exposure gain control signal, ES exposure timing control signal, EG / ES exposure control signal (exposure gain control signal / exposure timing control signal), EG (1) first exposure gain control signal for normal shooting, ES (1) First exposure timing control signal for normal shooting, EG (1) / ES (1) First exposure control signal for normal shooting, EG (2) Second exposure gain control for bracket shooting Signal, ES (2) Second exposure timing control signal for bracket shooting, EG (2) / ES (2) Second exposure control signal for bracket shooting, ET gradation Control signal, TS timing signal, VA imaging output, VB amplification output, VC A / D conversion output, VF feature image detection correction output, VG feature image detection enable / disable output, VH (feature image) feature image exposure correction value, VI Display output signal, VK feature image detection result.

Claims (10)

Recording means for preliminarily storing subject image data used for authentication of a plurality of subjects;
Imaging means for imaging a subject by bracket shooting for shooting a plurality of images with normal shooting or changing exposure, and outputting an image signal of the captured image;
Feature image detection availability determination means for detecting a predetermined feature image in the captured image from the image signal of the captured image and outputting at least a feature image detection availability signal capable of determining whether the feature image has been detected;
A feature image exposure correction value for correcting exposure by detecting a predetermined exposure correction signal in a predetermined feature image region including the feature image of the captured image is generated from the image signal and the feature image detection enable / disable signal. Exposure correction value detection means for
A first exposure control value for the first shooting of normal shooting or bracket shooting is calculated from the integrated value of the photometric window of the predetermined dimension, and bracket shooting is calculated from the first exposure control value and the characteristic image exposure correction value. A second exposure control value for the second and subsequent shootings with different exposures is calculated, and the exposure at the time of shooting is controlled based on the first exposure control value or the second exposure control value. Exposure control means for outputting an exposure control signal;
An authentication unit that determines whether or not authentication is possible using a captured image that has been shot with exposure controlled by the exposure control unit and image data for authentication stored in the recording unit ;
Equipped with a,
The authentication means includes
When the feature image is not included in the photometric window of the predetermined size, exposure is controlled using the exposure control signal based on the first exposure control value and the second exposure control value. Using the captured image and the image data for authentication stored in the recording means to determine whether authentication is possible,
In the case where the feature image is included in the photometric window having the predetermined size, a captured image obtained by controlling exposure using the exposure control signal based on the first exposure control value, and the recording unit To determine whether authentication is possible using the authentication image data stored in
An authentication apparatus characterized by that . The exposure correction value detecting means includes
Selecting one captured image from which the feature image recognized as an image can be optimally detected using the predetermined exposure correction signal and the feature image detection enable / disable signal from each captured image captured by the bracket. The authentication apparatus according to claim 1, wherein the authentication apparatus is characterized. The feature image detection availability determination means includes
The authentication according to claim 1 or 2, wherein a predetermined feature image in a plurality of captured images obtained by bracket shooting is detected, and a feature image detection availability signal is output for the captured image from which the predetermined feature image is detected. apparatus. In the characteristic image detection availability determination means, a luminance signal, color difference signal, contrast difference signal, RGB signal obtained by converting the image signal for each photometric window having a predetermined size formed by dividing one screen into a horizontal direction and a vertical direction. The authentication apparatus according to any one of claims 1 to 3, wherein at least one of the signals is used. The image signal is
The authentication apparatus according to any one of claims 1 to 4, wherein the authentication apparatus is directly output to the feature image detection availability determination unit, the exposure correction value detection unit, and the authentication unit. The feature image detection availability determination unit, the exposure correction value detection unit, and the authentication unit are connected to a system control unit that controls the entire system through a bus,
The image signal is
The authentication apparatus according to claim 1, wherein the authentication apparatus outputs the characteristic image detection availability determination unit, the exposure correction value detection unit, and the authentication unit via the bus. Display means for displaying an image based on the image signal, and external sensor means for detecting an image displayed on the display means,
The authentication apparatus according to claim 1, wherein the feature image detection availability determination unit detects a predetermined feature image in the captured image when the external sensor unit detects an image. . A table of a plurality of gradation conversion characteristics is held, and gradation conversion means for correcting gradation of the digitized image signal and outputting the corrected image signal is provided.
The exposure correction value detection means uses a contrast difference signal as the predetermined exposure correction signal,
The authentication apparatus according to claim 1, wherein the exposure control unit outputs a gradation control signal for switching a table of gradation conversion characteristics of the gradation conversion unit. The authentication according to claim 1, wherein the exposure correction value detection unit detects a predetermined exposure correction signal in a feature image region in which the predetermined feature image region is enlarged. apparatus. Network means for connecting the imaging means to a network,
The authentication device according to claim 1, wherein the authentication unit is connected to another unit via the network unit, and performs authentication via the network via the network unit. .

Images