JP5385737B2 - Focus detection apparatus and method - Google Patents

Description

  The present invention relates to an automatic focus detection technique.

  Conventionally, a TTL (Through The Lens) type autofocus (AF) method has been proposed. In this AF method, a high-frequency component of an imaging signal is extracted while moving the focus lens, and the focal point is detected by searching for the position of the focus lens (the peak of the mountain) where the high-frequency component is maximum. Hereinafter, this type of AF is referred to as an imaging signal AF. This imaging signal AF method can be realized at low cost because there is no need for a mechanical member or the like for automatic focus detection. In addition, since the degree of focus is determined from the imaging signal, there are features such as high accuracy and no change over time. In particular, in imaging with a high-pixel imaging device, since there is a high demand for focus accuracy, the imaging signal AF that is not affected by changes over time of the mechanical member is effective.

  On the other hand, this imaging signal AF method scans the drive range of the focus lens from end to end in order to determine the peak of the mountain, that is, the in-focus point, or grasps the shape of the mountain, or so-called mountain climbing operation Therefore, there are the following drawbacks. First, when the shape of a mountain is grasped by scanning, an operation of defocusing and seeing is visible, so that there is a disadvantage that it is inappropriate in a state where a movie is taken or displayed on a monitor. In addition, hill-climbing movement is generally used for video shooting and preview operation on a monitor. If the focus lens is far away from the focal point and starts moving, Starts from a flat part. In such a case, it is difficult to determine the direction of the top of the mountain, and if you make a mistake in the direction of the top of the mountain, you can go back to the opposite end from the top of the mountain and return to the image until it is in focus. Becomes unsightly and takes a lot of time to focus.

  On the other hand, TTL and external autofocus (hereinafter referred to as direct measurement AF) methods using an infrared triangulation method and a pupil division phase difference detection method have also been proposed. In this direct measurement AF method, it is possible to obtain information related to the distance to the subject directly, and since it does not require a scanning operation or a hill climbing operation for detecting the position of the focal point, the focal point can be determined at high speed. There are features. However, since a mechanical member or the like for focus detection is required, the cost is high, and measurement is performed using a system different from the imaging system.

  In addition, hybrid AF has been proposed that realizes high-precision and high-speed autofocus by combining the two AF methods described above and configuring a system so as to compensate for each other's drawbacks.

  However, when the imaging system for the imaging signal AF and the imaging system for the direct measurement AF are arranged at optically different positions, there is a problem that parallax occurs between the two imaging systems. There is a point. Due to this parallax, the measurement target area of the direct measurement AF may be an area that does not include a desired subject. In order to solve such a problem, a technique has been proposed in which an imaging target area for the imaging signal AF and a measurement target area of the direct measurement AF overlap (for example, refer to Patent Document 1).

  On the other hand, there is a demand for focusing on a moving object in an imaging screen as a main subject. In response to this demand, a technique has been proposed in which a plurality of distance measuring sensors corresponding to fixed positions in the screen are provided and an appropriate distance measuring sensor is selected in accordance with the movement of the moving object in the screen ( For example, see Patent Document 2).

JP 2008-26804 A JP 2001-194578 A

  However, when the subject is a moving object, if the external measurement direct measurement AF that can measure the distance at high speed is applied to the subject, the above-mentioned parallax becomes a problem and the direct measurement AF is successfully applied. There is a problem that it cannot be done.

  The present invention has been made in view of the above problems, and an object thereof is to enable continuous focusing on a moving object at high speed by direct measurement AF of an external measurement method.

In order to achieve the above object, a focus detection apparatus of the present invention includes a moving object detection unit that detects a moving object included in an image of an imaging signal obtained from an imaging unit, a plurality of measurement areas, and is included in each measurement area. Acquiring information on the distance to the subject to be captured at a preset cycle, acquisition means disposed at a position optically different from the imaging means, and one or more of the plurality of measurement areas corresponding to the moving object Selection means for selecting a measurement area, and a control signal for driving a focus lens so that the moving body is in focus based on information on the distance to the moving body obtained from the measurement area selected by the selection means Generating means for generating the information, and the selecting means relates to the information about the distance obtained from the selected measurement area and the distance obtained from each measurement area in the next cycle. That information and is based on the difference of the distances respectively, reselect the measurement area.

  According to the present invention, it is possible to continue focusing on a moving object at high speed by the direct measurement AF of the external measurement method.

1 is a block diagram showing a schematic configuration of a camera in an embodiment. Explanatory drawing of a structure and operation | movement principle of the direct measurement AF module in embodiment. The figure showing the example of installation layout of a direct measurement AF module. The figure explaining the parallax of an imaging optical system and AF optical system. 6 is a flowchart for explaining an automatic focus detection operation in the embodiment. 6 is a flowchart for explaining an automatic focus detection operation in the embodiment. The flowchart explaining the selection process of the measurement area with respect to a moving body area. The figure which shows the overlapping condition of the moving body and measurement area in each distance. The flowchart explaining the area selection method when there is no moving body in an image. The figure explaining the overlapping condition of a measurement area and a center area. The figure explaining the change of the cross rate according to subject distance. 5 is a flowchart for explaining measurement area setting processing according to the first embodiment. The figure which shows the movement of the imaging surface direction of the moving body in 1st Embodiment, and how to overlap a measurement area. 9 is a flowchart for explaining measurement area setting processing according to the second embodiment. The figure which shows the movement of the optical axis direction of the moving body in 2nd Embodiment. The figure which shows how the moving body which is moving to the optical axis direction in 2nd Embodiment, and the measurement area overlap. Explanatory drawing when using a plurality of measurement areas simultaneously. Explanatory drawing which shows how to allocate another measurement area.

  The best mode for carrying out the present invention will be described below in detail with reference to the accompanying drawings.

  FIG. 1 is a block diagram showing a configuration of a camera (imaging apparatus) equipped with an automatic focus detection apparatus according to an embodiment of the present invention. The camera here is a generic term for a so-called video camera, digital still camera, or the like that captures moving images and still images and records them on various media such as a tape, a solid-state memory, an optical disk, and a magnetic disk. Each unit in the camera is connected via a bus 260, and each unit is controlled by the main CPU 251.

  The lens unit 201 includes a fixed first group lens 202, a zoom lens 211, a diaphragm 203, a fixed third group lens 221, and a focus lens 231. Through these optical members, a subject image is formed on the image sensor 241 and imaged. The zoom control circuit 213 changes the focal length by driving the zoom lens 211 via the zoom motor 212 in accordance with an instruction from the main CPU 251.

  The subject image formed on the image sensor 241 is photoelectrically converted, and the obtained electric signal is adjusted to the image signal TVS by the image signal processing circuit 242 and input to the AF signal processing circuit 234 and the moving object detection processing circuit 235. . The AF signal processing circuit 234 generates an evaluation value FV for controlling the imaging signal AF (mountain climbing AF) and an IFA signal indicating the degree of focus, and inputs the generated IFA signal to the focus control circuit 233. The moving object detection processing circuit 235 searches the image signal TVS for a portion corresponding to the moving object, and if found, inputs the position and size of the moving object part (moving object area) in the imaging signal to the focus control circuit 233. .

  On the other hand, the direct measurement AF detection module 230 provided outside the lens unit 201 calculates a signal P (X) (X represents a measurement area to be described later) that is information (distance information) related to the subject distance. To the focus control circuit 233. The calculation of the information regarding the subject distance is performed by detecting the phase difference between the two subject images formed on the sensor 239 as the phase difference detector through the pupil division optical system 238 for direct measurement AF. In the focus control circuit 233, the focus motor 232 is controlled by a control signal based on the signal P (X) obtained as described above, the evaluation value FV, the degree of focus IFA, and the position and size of the moving object based on the moving object detection result. Then, the focus lens 231 is driven. Thereby, autofocus (AF) is realized.

  The image signal TVS adjusted by the imaging signal processing circuit 242 is temporarily stored in the RAM 254. The image signal TVS stored in the RAM 254 is compressed by the image compression / decompression circuit 253 when recorded image recording is instructed, and is recorded in the image recording medium 257. In parallel with this, the image signal TVS stored in the RAM 254 is reduced or enlarged to an optimum size for display by the image processing circuit 252 and displayed on the monitor display 250. Thereby, at the time of moving image shooting, the shot image can be fed back to the photographer in real time. In other words, in the case of moving image shooting, the moving image remains on the liquid crystal screen. Further, at the time of still image shooting, it is possible to confirm the shot image by displaying the shot image on the monitor display for a predetermined time immediately after shooting one captured image.

  The operation switch 256 is used for an operation for the user to give an instruction. Reference numeral 259 denotes a power supply battery, which is subjected to appropriate power management by the power management circuit 258 so that stable power supply is performed for the entire camera.

  Prior to these operations, when the camera is started from the OFF state, the program stored in the flash memory 255 is loaded into a part of the RAM 254. The main CPU 251 performs the above-described operation according to the program loaded in the RAM 254.

  Next, the configuration and operation principle of the direct measurement AF detection module 230 provided outside the lens unit 201 in the present embodiment will be described with reference to FIG.

Reference numeral 301 denotes a subject, and 239L and 239R are left and right sensor arrays (hereinafter referred to as “line sensors”) in which photoelectric conversion sensors that constitute the sensor 239 are arranged side by side. The left line sensor 239L and the right line sensor 239R are provided with a left AF sensor lens 238L and a right AF sensor lens 238R, respectively, constituting a pupil division optical system 238 for direct measurement AF. The light flux from the subject 301 passes through the respective optical paths and enters the left line sensor 239L and the right line sensor 239R via the left AF sensor lens 238L and the right AF sensor lens 238R. Here, the left AF sensor lens 238L, the right AF sensor lens 238R, the left line sensor 239L, and the right line sensor 239R are used as the direct measurement AF detection module 230. The distance l from the direct measurement AF detection module 230 to the subject 301 is the base line length B, the focal length f of the direct measurement AF detection module 230, and the phase difference of the left line sensor 239L with respect to the right line sensor 239R as n. ,
l = B × f / n

Can be obtained. Based on this result, the amount of extension of the focus lens 231 to the in-focus position can be obtained as a function of the distance l or the phase difference n as information on the subject distance.
Note that the left line sensor 239L and the right line sensor 239R are divided into three, and are defined as measurement areas L (left), C (center), and R (right) from the left side. A distance l as information on the subject distance is obtained for each measurement area, and is input to the focus control circuit 233 as a signal P (X) (X is any one of L, C, and R). Note that the left AF sensor lens 238L and the right AF sensor lens 238R are integrally molded so that the baseline length B is substantially constant (predetermined value) in each camera. This is because it is possible to prevent the base line length from shifting due to disassembly or impact from the outside by performing integral molding (single package). In FIG. 2, the distance between the centers of the two lenses (the mutual distance between the two sensors behind the lens) is a certain value.

  FIG. 3 is a diagram showing an installation layout example of the direct measurement AF detection module 230. FIG. 3 is a front view of the camera as viewed from the subject side, and a direct measurement AF detection module 230 as an AF optical system is installed on the left side of a lens unit 201 as an imaging optical system. This is merely an example, and the direct measurement AF detection module 230 may be installed on the right side or obliquely above / below, but here, it is installed on the left side of the lens unit 201 as shown in FIG. An example will be described. When the direct measurement AF detection module 230 is installed as shown in FIG. 3, parallax occurs between the imaging optical system and the AF optical system with respect to a change in distance.

  FIG. 4 is a diagram for explaining the parallax in the layout shown in FIG. Here, the area on the subject side reflected in the image sensor 241 and the sensor 239 (both the left line sensor 239L and the right line sensor 239R) is a short distance (for example, 1 m), a medium distance (for example, 3 m), and a long distance (for example, 100 m). ) At a short distance, the image of the area 502 on the subject side is projected onto the imaging area of the image sensor 241, and the image of the area 512 is projected onto the measurement areas L (left), C (center), and R (right) of the line sensor 239. Is done. Further, at an intermediate distance, the image of the region 503 is projected onto the imaging area of the image sensor 241 and the image of the region 513 is projected onto the measurement areas L (left), C (center), and R (right) of the line sensor 239, respectively. . At a long distance, the area 504 is projected onto the imaging area of the imaging device 241, and the area 514 is projected onto the measurement areas L (left), C (center), and R (right) of the line sensor 239.

  As shown in FIG. 4, at a short distance (1 m), the measurement area region 512 is located on the right side with respect to the imaging area 502. In this case, the measurement areas L and C of the measurement area region 512 overlap with the imaging area 502, and the measurement area R does not substantially overlap with the imaging area 502. At a medium distance (3 m), the measurement area region 513 and the center of the imaging area 503 are substantially coincident with each other, and the measurement areas L, C, and R all overlap the imaging area 503. At a long distance (100 m), the measurement area region 514 is located from the left with respect to the imaging area 504. In this case, the measurement areas C and R in the measurement area region 514 overlap with the imaging area 504, and the measurement area L does not substantially overlap with the imaging area 504.

  Note that the description regarding the parallax described above with reference to FIG. 4 assumes that the zoom lens 211 is fixed at a certain position and the angle of view of the captured image is fixed. If the focal length is changed by moving the zoom lens 211, the angle of view of the captured image changes. In this case, the way in which the imaging area and the measurement area overlap is changed. Therefore, in the camera according to the present embodiment, it is assumed that information on how the imaging area and the measurement area overlap is held as data in the flash memory 255 with respect to the angle of view of each position of the zoom lens 211.

<First Embodiment>
5A and 5B are flowcharts for explaining the automatic focus detection operation in the camera having the above-described configuration in the first embodiment. The operation is controlled by the main CPU 251 unless otherwise specified. In the present embodiment, control is performed by switching the following three modes as the AF mode. The first AF mode is a “normal AF mode” in which the in-focus state is maintained by the imaging signal AF after focusing by direct measurement AF. The “normal AF mode” is a mode in which the direct measurement AF is roughly focused and fine adjustment is performed with the imaging signal AF. The second is a “moving object high-speed AF mode” in which measurement AF is continuously performed on the moving object and the moving object is focused at high speed. The third is a “pre-moving object AF mode” which is a mode in which the moving object is once focused with the imaging signal AF. Since it is necessary to detect the moving object part by image processing before performing the “pre-moving object AF mode” and “moving object high speed AF mode”, the moving object part is focused by the imaging signal AF after the moving object part is detected. It is a mode to do. The “moving body high-speed AF mode” is performed after the “pre-moving body AF mode”. The distance from the focusing lens position to the moving object is calculated, the measurement area (R, C, L) in the direct measurement AF that overlaps with the moving object is identified according to the distance, and AF is performed by the direct measurement AF using the area. Mode.

  When the automatic focus detection operation is started, first, 1 is substituted into AFMODE representing the AF state (S11). Note that AFMODE has the following three states. If AFMODE = 1, focusing processing is performed using direct measurement AF, and focusing processing is performed using imaging signal AF when AFMODE = 2. Also, if AFMODE = 3, assuming that the in-focus state has been completed, it is monitored whether there is any change in the focus determination value. If there is a change, AFMODE = 1 or AFMODE = 2 is set for the focus process. Processing to migrate. Here, AFMODE = 1 is an operation for performing direct measurement AF, and corresponds to a case where the degree of focus in the “moving object high-speed AF mode” or the “normal AF mode” is low. AFMODE = 2 corresponds to the contrast AF operation when the degree of focus is high in the “pre-moving body AF mode” or the “normal AF mode”. Also, AFMODE = 3 corresponds to the case where the focus lens does not move without performing the AF operation regardless of which “moving body high speed AF mode”, “pre-moving body AF mode”, and “normal AF mode”. To do.

  Subsequently, image signal processing by the imaging signal processing circuit 242, AF signal processing by the AF signal processing circuit 234, and phase difference detection processing based on the output of the sensor 239 are performed in synchronization (S 12). Then, the IFA signal indicating the evaluation value FV and the degree of focus is received from the AF signal processing circuit 234, and the signals P (L), P (C), and P (R) indicating information related to the subject distance from the direct measurement AF detection module 230. Is acquired (S13). If a moving object is detected at this time, an evaluation value FV and an IFA signal for the moving object area are also acquired. In S14, the measurement area of the sensor 239 used for direct measurement AF is set. The processing performed here will be described later in detail.

  In S15, the moving object detection processing circuit 235 performs a moving object detection process on the image signal TVS received from the imaging signal processing circuit 242. In this process, it is determined whether or not there is a moving object that is a subject, and if there is a moving object, the position and size of the moving object (moving object area) in the screen are calculated. If a moving object exists (YES in S16), the process proceeds to S17. If a moving object does not exist (NO in S16), the process proceeds to S27.

  In S17, it is determined whether or not the moving object AF_FLAG indicating the “moving object high speed AF mode” is ON. When the moving object AF_FLAG is ON, the measurement area for performing the moving object high speed AF has already been selected and the moving object high speed AF is being performed, and thus the process proceeds to S30 in FIG. 5B. On the other hand, if it is determined that the moving object AF_FLAG is OFF, the process proceeds to S18, and it is determined whether or not the IFA signal indicating the degree of focus of the moving object area is greater than a predetermined value IFAd. This is a determination of whether or not the moving object area is properly focused. If it is determined that the moving object area is not properly focused, the process proceeds to S25. In this case, the moving object is detected, but the focusing is not properly performed on the moving object area. Therefore, the pre-moving object AF_FLAG is turned ON in order to shift to the “pre-moving object AF mode” in which the moving object is temporarily focused only by the imaging signal AF (S25). Then, 2 is substituted into AFMODE to perform the imaging signal AF (S26), and the process proceeds to S30 in FIG. 5B. That is, the user wants to enter the “moving body high-speed AF mode” that continues to match the moving object only by direct measurement AF, but does not obtain the distance information to the moving object. Enter "mode".

  On the other hand, when it is determined that focusing on the moving object area is appropriate (YES in S18), the process proceeds to S19, and measurement area selection processing for the moving object area for performing moving object high speed AF is performed.

  FIG. 6 is a flowchart showing an example of an algorithm for measurement area selection processing for a moving object area in the present embodiment, which is performed in S19.

  First, the position and size of the moving object area obtained from the moving object detection processing circuit 235 are acquired (S61). Next, since the moving object is appropriately focused at present, the distance from the camera to the moving object (moving object distance) is calculated from the current position of the zoom lens 211 and the position of the focus lens 231 ( S62). Then, based on the moving object distance, the arrangement of the measurement area with respect to the imaging area is determined for the subject at the moving object distance (S63). As described above with reference to FIG. 4, the measurement area and the imaging area overlap depending on whether the moving object distance is a short distance (1 m), a medium distance (3 m), or a long distance (100 m). Because it changes.

  Next, the measurement area that overlaps the moving object area is selected (S64). In the example shown in FIG. 7, when the moving object distance is 1 m, the measurement area L overlaps the moving object area 700 most, so the measurement area L is selected. When the moving object distance is 3 m, the measurement area C overlaps with the moving object area 700 most, so the measurement area C is selected. When the moving object distance is 100 m, the measurement area R overlaps with the moving object area 700 most, so the measurement area R is selected. Then, the degree of overlap between the moving object area and the selected measurement area is compared with a previously held threshold value (S65). If the degree of overlap is greater than or equal to the threshold value, the measurement area corresponding to the measurement area selected in S64 is substituted into X as the measurement area to be used, and if it is less than the threshold value, the measurement selected in S64. Returning to S19 of FIG. 5, assuming that the measurement area corresponding to the area is not used.

  With the above processing, when there is a moving object in the captured image, it is possible to select a measurement area for performing measurement AF directly on the moving object part.

  Next, in S20 of FIG. 5A, it is determined whether or not an appropriate measurement area has been selected for the moving object in S19. If it can be selected, the moving object AF_FLAG is turned ON (S21), and the pre-moving object AF_FLAG is turned OFF. (S22). In S23, it is determined whether or not AFMODE = 2, which is not used during moving body high speed AF. If AFMODE is 2, it is set to 1 (S24). If AFMODE is not 2, it is in a state where focusing is performed by moving body high speed AF (AFMODE = 1) or in a state where focusing is performed by moving body high speed AF (AFMODE = 3), so AFMODE is not changed. Proceed to S30 of FIG. 5B.

  If it is determined that there is no moving object (NO in S16), and if it is determined that an appropriate measurement area for the detected moving object cannot be selected (NO in S20), the process proceeds to S27. In S27, the measurement area located at the center of the imaging region as much as possible is selected from the measurement areas L, C, and R regardless of the moving object. FIG. 8 is a flowchart showing a specific algorithm of the measurement area selection process performed in S27 of FIG. 5A, and FIG. 9 is a cross rate showing the overlap between the center area 901 set in the imaging area 900 and the measurement area 902. It is a figure explaining the calculation method.

In FIG. 9, a indicates the width of the measurement area C, and b indicates the width of the portion where the central area 901 and the measurement area C overlap. The cross rate CC (subject distance) corresponding to the subject distance in the measurement area C in FIG. 9 is expressed as follows.
CC (subject distance) = b / a

  The cross ratio LC (subject distance) and RC (subject distance) corresponding to the subject distance in each of the measurement area L and the measurement area R can be similarly obtained.

  FIG. 10 is a graph showing the change of the cross rate depending on the subject distance. In the camera of the present embodiment, parameters corresponding to this graph are held in advance. The horizontal axis is the subject distance, and the vertical axis is the cross rate. The cross rate LC of the measurement area L is maximum near 1 m, the cross rate CC of the measurement area C is maximum near 3 m, and the cross rate RC of the measurement area R is maximum near 100 m. Further, the cross rate CC of the measurement area C is larger than the cross rate LC of the measurement area L with the subject distance D1 as a boundary, and the cross rate RC of the measurement area R is greater than the cross rate CC of the measurement area C with the subject distance D2 as a boundary. Will be bigger.

  The algorithm of the measurement area selection process will be described with reference to FIG. 8 based on the description of FIGS.

  First, when it is determined in S71 that the distance D (C) corresponding to the signal P (C) of the measurement area C is larger than D1 and smaller than D2, the process proceeds to S74 and the measurement area C is selected. If not (NO in S71), the process proceeds to S72. As described above, the measurement area C is first checked because the measurement area C has priority over other measurement areas. In the case of video shooting, the main subject generally intended by the photographer is displayed on the screen. This is because they often come to the center of the city.

  In S72, it is checked whether or not the distance D (L) corresponding to the signal P (L) in the measurement area L is equal to or less than D1, and if it is equal to or less than D1, the process proceeds to S75 and the measurement area L is selected. On the other hand, if the distance D (L) corresponding to the signal P (L) is larger than D1 in S72, the process proceeds to step S73.

  In S73, it is checked whether the distance D (R) corresponding to the signal P (R) in the measurement area R is equal to or greater than D2, and if it is equal to or greater than D2, the process proceeds to S76 and the measurement area R is selected. On the other hand, if the distance D (R) corresponding to the signal P (R) is less than D2 in S73, the process proceeds to S74 and the measurement area C is selected. Here, the measurement area C is selected because the measurement area C has priority over other measurement areas. Therefore, in the flowchart of FIG. 8, the measurement area L is determined in S72 and the measurement area R is determined in S73, but this order may be reversed. That is, the measurement area R may be determined in S72 and the measurement area L may be determined in S73. In that case, the measurement area R has priority over the measurement area L. The priority given to the measurement area C in this manner means that the subject distance 3m is given priority in the present embodiment, as can be seen from FIG.

  When the selection of the measurement area when no moving object is detected as described above, the process returns to step S27 in FIG. 5A. In S28, OFF is assigned to the moving object AF_FLAG to turn OFF the moving object high speed AF, and OFF is also assigned to the pre-moving object AF_FLAG.

  Next, the process after S30 of FIG. 5B is demonstrated. First, in S30, the value of AFMODE is determined. Immediately after the camera is operated, AFMODE is 1, so the process proceeds to S31. In S31, the lens driving direction and speed of the focus lens 231 are determined based on the signal P (X) obtained from the measurement area selected in S19 or S27 and the current position of the focus lens 231. In other words, the direction and speed are set so that the position of the focus lens 231 approaches the in-focus position corresponding to the signal P (X), the speed is high when the distance is long, and the speed is low when the distance is close. decide. Next, in S32, it is determined whether or not the in-focus position corresponding to the signal P (X) matches the current position of the focus lens 231. When it is determined that the in-focus state is obtained, it is determined to stop driving the focus lens 231 (S33). If the moving object AF_FLAG is ON (YES in S34), 3 is assigned to AFMODE (S35), and if the moving object AF_FLAG is OFF (NO in S34), 2 is assigned to AFMODE (S36). In this embodiment, in the normal AF mode in which the moving body high speed AF is not performed, the imaging signal AF is focused after the direct measurement AF, but the moving body high speed AF is performed. This is because the moving AF is always focused directly on the moving object by measurement AF. Then, it progresses to S49 and stops a motor. If it is determined in S32 that the focus is not achieved, the process proceeds to S49, and the focus lens 231 is driven in the direction determined in S31 at the determined speed. After S49, the process returns to S12 in FIG. 5A, and thereafter, the above-described processing is repeated in synchronization with the readout cycle of the imaging device (imaging signal processing cycle).

  If AFMODE is 2 in S30, the process branches to S37. In S37, the pre-moving object AF_FLAG is checked to determine whether or not the pre-moving object AF mode is currently set. If it is determined that the pre-moving body AF mode is not selected, the process proceeds to S38. In S38, it is determined whether the evaluation value FV acquired in S13 is increasing or decreasing compared to the evaluation value FV acquired last time. Although not described in the algorithm, the previous evaluation value FV is assumed to be held in the RAM 254, for example. When the evaluation value FV is increasing, it progresses to S49, and when it is not increasing, it progresses to S39. In S39, it is determined that the driving direction of the focus lens 231 is reversed, and in S40, it is determined whether or not it is a decrease after passing the peak of the evaluation value FV. If it is determined in S40 that the peak has not been passed, the process proceeds to S49. If it is determined that the peak has been passed, 3 is substituted for AFMODE in S41, and then the process proceeds to S49.

  If it is determined in S37 that the current pre-moving object AF mode is selected, the process proceeds to S42, where it is determined whether or not the evaluation value FV in the moving object area has increased from the previous evaluation value FV. If so, the process proceeds to S39 to perform the above-described processing. If it has increased, the process proceeds to S49. In this way, by using the evaluation value FV of only the moving object region for determination, the moving object region portion is focused by the imaging signal AF.

  When AFMODE is determined to be 3 in S30, the process proceeds to S43, and the focus lens 231 is returned to the peak position of the evaluation value FV and stopped (determined to be in focus). In S44, it is monitored how much the evaluation value FV has changed from the level of the evaluation value FV at the position (focus position) of the focus lens 231 at which the evaluation value FV reaches a peak. In parallel, it is also monitored whether or not the focus position corresponding to the signal P (X) has changed. If not changed, the process proceeds to S49, and the focused state is maintained. On the other hand, if it has changed, it is determined whether or not the IFA signal is smaller than the in-focus threshold value IFAth (determination of whether it is blurring or blurring) in S45. Then, the process proceeds to S46. In S46, 1 is substituted into AFMODE and the process proceeds to S49 (restart from AFMODE1). On the other hand, if the IFA signal is greater than or equal to IFAth in S45, it is determined that the amount of blur is small, and the process proceeds to S47. If the moving object AF_FLAG is ON, the process proceeds to S46. If it is OFF, the process proceeds to S48 and 2 is assigned to AFMODE, and then the process proceeds to S49. In this case, restart from AFMODE2.

  FIG. 11 is a flowchart showing the measurement area setting process of the sensor 239 used for the direct measurement AF in S14 in the flowchart of FIG. 5A in the first embodiment.

  Here, the operation assumed in the first embodiment will be described with reference to FIG. For example, it is assumed that there is a moving object at a medium distance (for example, 3 m) as shown in FIG. In this case, the movement from the region 1801 to the region 1802 of the imaging area 503 in FIG. 12A corresponds to the following operation. That is, as shown in FIG. 12B, from the state in which the distance can be correctly measured in the measurement area C of the sensor 239 corresponding to the measurement area C, the state in which the distance can be correctly measured in the measurement area R of the sensor 239 corresponding to the measurement area R. It corresponds to moving to. The first embodiment aims to change the measurement area appropriately in such a case.

  First, in S101 of FIG. 11, it is determined whether or not the moving object AF_FLAG is currently ON. If it is ON, the process proceeds to S102, and if it is OFF, the process is terminated as it is, and the process proceeds to S15 in FIG. 5A.

  In S102, it is determined whether or not a sudden change has occurred by looking at the temporal history of P (X) (X represents the measurement area), which is a distance measurement result of the moving object area performing the moving object AF. To do. Here, it is assumed that a threshold value for change determination is held in advance. If there is a sudden change here, the process proceeds to S103, and if there is no change, the process ends, and the process proceeds to S15 in FIG. 5A. In S103, the absolute value of the difference between the distance measurement data P (X) before the change of the measurement area X currently used for measurement and the current distance measurement data P (Y) of the measurement area Y other than the measurement area X is obtained. It is calculated and it is determined whether this value is smaller than the threshold value Pth. If it is determined that the moving object has been measured in the measurement area X, it is highly likely that the moving object has moved to the measurement area Y. Therefore, the measurement area is changed to Y (S104), and the process returns to FIG. 5A.

  When it is determined that the size is large in S103, the process proceeds to S105, and the same determination process is performed for the measurement areas Z other than the measurement areas X and Y. In S105, the absolute value of the difference between the distance measurement data P (X) before the change of the measurement area X currently used for measurement and the current distance measurement data P (Z) of the measurement area Z is calculated. It is determined whether this value is smaller than the threshold value Pth. If it is determined that the measurement area is small, the measurement area is changed to Z (S106). If it is determined in S105 that it is large, it is determined that the area where the moving object can be detected is lost. become.

  As described above, according to the first embodiment, a moving object moving in parallel with the imaging screen can be focused at high speed by the direct measurement AF method. Even when the measurement area where the moving object can be measured is lost, the normal AF mode can be returned to and appropriate processing can be performed.

<Second Embodiment>
FIG. 13 is a flowchart showing the measurement area setting process of the sensor 239 used for the direct measurement AF in S14 in the flowchart of FIG. 5A in the second embodiment. This process is performed instead of the process of FIG. 11 described in the first embodiment. Since other than this is the same as in the first embodiment, the description is omitted.

  The operation assumed in the second embodiment will be described with reference to FIGS. In addition, the same reference number is attached | subjected to the structure similar to FIG. Here, as shown in FIG. 14, consider a case where a moving object approaches the camera perpendicularly. For example, assuming that the moving object is imaged at a middle distance (3 m) and a short distance (1 m) so that the moving object is located at the center of the screen, the moving object is at the same position on the imaging screen as shown in FIG. It seems that only the size has changed. However, it can be seen that the appropriate measurement areas are different as shown in FIG. As shown in FIG. 15A, the measurement area C is appropriate at 3 m, and as shown in FIG. 15B, the measurement area L is appropriate at 1 m. The purpose of the second embodiment is to set the measurement area of the sensor 239 appropriately corresponding to the measurement area in such a case.

  First, in S201 of FIG. 13, it is determined whether or not the moving object AF_FLAG is currently ON. If it is ON, the process proceeds to S202. If it is OFF, the process is terminated, and the process proceeds to S15 in FIG. 5A.

  In S202, the velocity in the direction perpendicular to the imaging screen of the moving object is obtained from the measurement history of the measurement distance data P (X) in the current measurement area X. Next, in S203, it is predicted how the measurement area appropriate for the moving object will change in the future based on the current measurement area, the distance to the current subject, and the speed of the subject.

For example, consider a case where the subject distance corresponding to the signal P (X) from the measurement areas L, C, R changes in time series as 1) → 2) → 3) below.
1) L: 10m, C: 3m, R: 7m
2) L: 10m, C: 2.5m, R: 7m
3) L: 2m, C: 10m, R: 7m

  In this case, the moving body is approaching in 1) and 2), and it can be predicted that the appropriate area will shift from C to L in the time indicated in 3). In this manner, an appropriate measurement area can be predicted in advance from the moving speed of the moving object in the vertical direction with respect to the imaging screen.

  In this case, when the speed of the moving body is constant speed movement, deceleration movement or acceleration movement, it may be calculated by an appropriate calculation formula according to the difference in the angle of view depending on the position of the zoom lens, or lookup The table may be referred to. Thus, it is assumed that an appropriate measurement area for a moving object is predicted in S203.

  In subsequent S204, it is determined whether or not to change the measurement area from the current time and the prediction calculated in the past S203. When the measurement area calculated in the past S203 and the current predicted measurement area are different, the process proceeds to S205, the measurement area X is changed to the appropriate measurement area obtained in S203, the process is terminated, and the process shown in FIG. Return to S14. On the other hand, if it is determined in S204 that they are the same, the process returns to S14 in FIG. 5A without changing the measurement area.

  As described above, according to the second embodiment, it is possible to perform high-speed focusing using the direct measurement AF method even for a moving object that moves perpendicularly to the imaging screen.

  The first embodiment and the second embodiment described above may be used in combination.

  In the first and second embodiments described above, the case where one measurement area is assigned as a distance measurement area for measuring the distance to the moving object has been described. However, one or more measurement areas may be used. You may make it allocate the above area. Depending on the angle of view, the measurement area of direct measurement AF may be arranged as 2302 with respect to the imaging area as shown in FIG. In this case, it is possible to perform measurement with higher accuracy by using both areas C and R of the measurement area 2302 with respect to the moving object area 2301.

  In the first and second embodiments described above, the measurement area of the sensor 239 of the direct measurement AF detection module 230 is divided into three areas L, C, and R. However, the pixel sensors that the L, C, and R have, respectively. It is good also as a structure which makes a measurement area combining freely. For example, as shown in FIG. 17, it is assumed that each of the areas L, C, and R of the measurement area 2502 is composed of 40 pixel sensors. In contrast to this arrangement, when the moving object area 2501 is arranged, it is possible to perform measurement with higher accuracy if the sensor area is configured by assigning pixels to the moving object area 2501 without excess or deficiency. In the example of FIG. 17, pixel selection such as “10 to 40 pixels in area C and 1 to 15 pixels in area R” is performed, and the pixel group is newly set as an appropriate measurement area for moving objects.

  In the first and second embodiments described above, it is assumed that there is a single moving object to be focused. However, when there are a plurality of moving objects, priorities are assigned according to size, position on the screen, and the like. It is also possible to carry out by selecting one.

  In the first and second embodiments described above, the moving object detection is always performed, and the measurement AF is directly performed only on the moving object. On the other hand, even if the detected moving object part does not move with respect to time, the part of the image currently acquired from the luminance information of the already detected moving object is recognized as a moving object. The measurement AF may be directly performed on the portion.

Claims (6)

Moving object detection means for detecting a moving object included in the image of the imaging signal obtained from the imaging means;
An acquisition unit that includes a plurality of measurement areas, acquires information about a distance to a subject included in each measurement area at a preset period, and is arranged at a position optically different from the imaging unit;
A selection means for selecting one or more measurement areas corresponding to the moving object among the plurality of measurement areas;
Generating means for generating a control signal for driving a focus lens so that the moving body is in an in-focus state based on information on the distance to the moving body obtained from the measurement area selected by the selection means; ,
The selection unit reselects a measurement area based on a difference between distances indicated by information on the distance obtained from the selected measurement area and information on a distance obtained from each measurement area in the next cycle. A focus detection apparatus. Moving object detection means for detecting a moving object included in the image of the imaging signal obtained from the imaging means;
An acquisition unit that includes a plurality of measurement areas, acquires information about a distance to a subject included in each measurement area at a preset period, and is arranged at a position optically different from the imaging unit;
A selection means for selecting one or more measurement areas corresponding to the moving object among the plurality of measurement areas;
Generating means for generating a control signal for driving a focus lens so that the moving body is in an in-focus state based on information on the distance to the moving body obtained from the measurement area selected by the selection means; ,
The selection means further predicts a distance to the moving object in the next cycle based on information on the distance obtained from the selected measurement area and information on a distance obtained from the selected measurement area in the next cycle. Then, the focus detection apparatus reselects the measurement area based on the predicted distance to the moving object. A calculation means for calculating a degree of focus from the imaging signal obtained from the imaging means;
The generation means generates the control signal based on the degree of focus before the measurement area is selected,
The selection means performs measurement based on the position of the focus lens that is in focus based on the degree of focus and the position of the moving body when selecting the measurement area from a state where the measurement area is not selected. The focus detection apparatus according to claim 1, wherein an area is selected.   The focus detection apparatus according to claim 3, wherein when the selection unit cannot reselect the measurement area, the generation unit generates the control signal based on the degree of focus. A moving object detecting step in which the moving object detecting unit detects a moving object included in the image of the imaging signal obtained from the imaging unit;
Acquisition means for acquiring information relating to the distance to the subject included in each measurement area at a preset cycle by an acquisition means arranged at a position optically different from the imaging means, including a plurality of measurement areas. Steps,
A selection step of selecting one or more measurement areas corresponding to the moving body from among the plurality of measurement areas;
A generating step for generating a control signal for driving the focus lens so that the moving body is in an in-focus state based on the information about the distance to the moving body obtained from the measurement area selected in the selection step. When,
Based on the difference in distance between the information related to the distance obtained from the measurement area selected in the selection step and the information related to the distance obtained from each measurement area in the next cycle, the selection means And a change step for reselecting the focus detection method. A moving object detecting step in which the moving object detecting unit detects a moving object included in the image of the imaging signal obtained from the imaging unit;
Acquisition means for acquiring information relating to the distance to the subject included in each measurement area at a preset cycle by an acquisition means arranged at a position optically different from the imaging means, including a plurality of measurement areas. Steps,
A selection step of selecting one or more measurement areas corresponding to the moving body from among the plurality of measurement areas;
A generating step for generating a control signal for driving the focus lens so that the moving body is in an in-focus state based on the information about the distance to the moving body obtained from the measurement area selected in the selection step. When,
Based on the information on the distance obtained from the measurement area selected in the selection step and the information on the distance obtained in the next period from the selected measurement area, the moving means in the next period is further selected. A prediction step to predict the distance to
The focus detection method, wherein the selection unit includes a change step of reselecting the measurement area based on the predicted distance to the moving object.

Images