JP2848749B2 - Feature image data extraction method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for extracting characteristic image data, and more particularly, to characteristics such as density data of a person's face used when copying a color original image onto a color copying material or a black and white copying material. The present invention relates to a method for extracting image data.

[0002]

2. Description of the Related Art The most noticeable part when viewing a photograph of a person is the face of the person. To create a high-quality photograph, the color of the face of the person must be adjusted appropriately. It is necessary to bake to a different color.

Conventionally, a face area in an original image of a color film is designated with a light pen to extract density data of a person's face, and the face color is appropriately printed based on the extracted density data. Exposure is determined. Such a technique is disclosed in JP-A-62-115430, JP-A-62-115431, and JP-A-62-1154.
No. 32, JP-A-62-189456, JP-A-62-189457, JP-A-63-138340
And JP-A-63-178222.

[0004] However, in the above-mentioned conventional technique, there is a problem that the printing operation takes a long time because the operator has to specify a face area with a light pen for each image. Further, since the operator must visually specify the face area, it is difficult to perform unmanned operation.

Therefore, a method of automatically extracting the face data of a person can be considered. For example, Japanese Patent Application Laid-Open No.
No. 56624, JP-A-52-156625,
JP-A-53-12330, JP-A-53-1456
No. 20, JP-A-53-145621, and JP-A-53-145622 describe the following method of extracting human face data by extracting skin color data. That is, the color original image is divided into a number of photometric points, and each photometric point is divided into R (red), G (green), B
Photometry is performed by decomposing the color into three colors (blue), and it is determined whether or not the color of each photometric point calculated from the photometric data is within the flesh color range. And
A cluster (group) of photometric points determined to be in the skin color range is used as face density data.

However, in this method, since the color within the skin color range is assumed as the density data of the face, parts other than the face having the skin color or the color similar to the skin color, such as the ground, a tree trunk, and clothes, are also included in the face. It is extracted as density data. Further, even when the same subject is photographed under the same conditions, the color of the photographed image differs depending on the film type, so that if the film type differs, the face density data may not be automatically extracted. Furthermore, if the color of the light source that illuminates the subject differs, the tint of the captured image differs (for example, an image captured using a fluorescent lamp as a light source becomes green). May not be extracted automatically.

In order to solve the above-mentioned problem caused by different light source colors, it is necessary to perform light source color correction and then extract photometric data in a flesh color range. As a light source,
Sunlight, fluorescent light, and tungsten light can be broadly classified. The color of sunlight varies depending on the season and time zone, and the color varies depending on whether the light is direct light or indirect light even in the same season and time zone. In addition, artificial lights such as fluorescent lamps have various colors with the diversification of products. Therefore, it is difficult to perform light source correction by specifying the light source type for each light source. Further, even if the light source correction can be completely performed, it is not possible to prevent extraction of a skin color or a portion similar to the skin color such as the ground or a tree trunk, and to cope with a case where the film type is different. Can not.

SUMMARY OF THE INVENTION The present invention has been made to solve the above problems, and it is possible to automatically extract only characteristic image data such as face data of a person from a color original image such as a negative film with high accuracy. It is an object of the present invention to provide a method for extracting characteristic image data that can be performed.

[0009]

According to the first aspect of the present invention, a color original image is divided into a large number of pixels and each pixel is separated into three colors of red light, green light and blue light. The color original image is divided for each color region having the same or similar hue value based on data obtained by the photometry, and the divided color region is in contact with the outer edge of the color original image. At least one color region other than the color region is selected, and data of the selected divided region is extracted as feature image data.

According to a second aspect of the present invention, a color original image is divided into a large number of pixels, and each pixel is decomposed into three colors of red light, green light and blue light, and photometry is performed. The color original image is divided into color regions having the same or similar hue value and saturation value based on the color regions, and the divided color regions and color regions other than the color region in contact with the outer edge of the color original image At least one color region is selected, and data of the selected divided region is extracted as feature image data.

In each of the above-mentioned inventions, when selecting an area, it is determined whether or not the divided area is a person's face, and the area determined to be a person's face is selected. It can be extracted as feature image data.

[0012]

According to the first aspect of the present invention, a color original image is divided into a large number of pixels, and each pixel is decomposed into three colors of red light, green light and blue light, photometry is performed, and a hue is obtained based on data obtained by photometry. The color original image is divided for each color region having the same or similar value. Therefore, one color region on the color original image includes pixels having the same or similar hue values within a predetermined range. The region of interest that represents the characteristics of the color original image is mainly near the center of the color original image, and therefore, for each of the divided color regions of the color original image, other than the color region that is in contact with the outer edge of the color original image If at least one color area is selected from among the color areas, the data of the selected color area represents the characteristic image data. Therefore, by selecting a color area other than the color area in contact with the outer edge of the color original image, Characteristic image data can be extracted.

If there is a change in the type of film or light source, a change over time, a difference in film development, or the like, the color of the color original image changes uniformly over the entire screen, but such a change in color occurs over the entire screen. Therefore, the range of the color region having the same or similar hue value of the color original image does not change even if the color tone changes. Therefore, in the present invention, data of a characteristic image can be extracted even if the color of a color original image changes due to a change in film type or light source type, a change over time, a film development difference, or the like.

If the hue of the characteristic image, which is the characteristic part of the image, is the same as or approximate to the hue of another part, dividing the original color image based on the similarity of only the hue values, In some cases, it cannot be distinguished from the site. Therefore,
According to the second aspect of the present invention, a saturation value is further introduced in addition to the hue value, and the original color image is divided for each color region having the same or similar hue value and saturation value. The characteristic image data is extracted by selecting at least one of the color regions other than the color region in contact with the outer edge of the color original image of the divided color region. In the present invention, since the hue value and the saturation value are used, the feature image data can be extracted even if the feature image and the part having the same or similar hue are mixed.

The most noticeable part when viewing a portrait is a person's face. Therefore, it is determined whether or not the divided region of the color original image is a person's face. It is preferable to extract the data of the area as feature image data. For example, the hue of an Asian person's face is similar to the skin color of the ground, trees, etc., but in most cases the saturation is different, so the hue value or hue value in contact with the outer edge of the original color image If the data is extracted from the color region excluding the color region having the same or similar saturation value, even the image in which the face, the ground, the tree, and the like are mixed is in contact with the outer edge such as the background of the original color image. The data of the face of the person can be extracted by excluding the ground and trees.

The data to be extracted as the characteristic image data may be other than the data of the face of a person.

[0017]

Embodiments of the present invention will be described below in detail with reference to the drawings. In this embodiment, the present invention is applied to an auto printer. As shown in FIG. 1, the automatic printer according to the present embodiment includes a transport roller 12 that transports a color negative film 10. A light source 1 is provided below the color negative film 10 transported by the transport rollers 12.
4. A color correction filter 16 such as a dimming filter and a diffusion box 18 are sequentially arranged. Above the negative film 10, the light transmitted through the negative film 10 is
A distribution prism 20 for distributing in the direction is arranged.
On one optical path distributed by the distribution prism 20, a projection optical system 22, a black shutter 23, and color paper (printing paper) 24 are sequentially arranged, and on the other optical path, a projection optical system 26 and a CCD image sensor. 28 are arranged in order. This CCD image sensor 28
The entire screen (one frame) of the negative film 10 is divided into a large number of pixels (for example, 256 × 256 pixels), and each pixel is
Photometry is performed by separating the color into three colors (red), G (green), and B (blue). The CCD image sensor 28 is connected to a 3 × 3 matrix circuit 34 for correcting the sensitivity of the CCD image sensor via an amplifier 30 for amplifying the output of the CCD image sensor and an analog-digital (A / D) converter 32. The 3 × 3 matrix circuit 34 is connected to an appropriate exposure amount calculation circuit 40 via a face extraction circuit 36 composed of a microcomputer storing a program of a routine described below, and calculates the average density of the entire screen. It is connected to an appropriate exposure calculation circuit 40 via an average density calculation circuit 38 for calculating. The appropriate exposure amount calculation circuit 40 is connected to the color correction filter 16 via a driver 42 for driving a color correction filter.

Next, the operation of this embodiment will be described. Light source 14
Are transmitted through the color correction filter 16, the diffusion box 18, and the color negative film 10, are distributed by the distribution prism 20, and are received by the CCD image sensor 28 via the projection optical system 26. At this time, the black shirt 23 is closed. By receiving the light, the CCD image sensor 28 divides the entire screen into a large number of pixels, decomposes each pixel into three colors of R, G, and B, performs photometry, and outputs a photometric data signal. The photometric data signal is amplified by an amplifier 30 and then converted into a digital signal by an A / D converter 32. The sensitivity of the image sensor is corrected by a 3 × 3 matrix circuit 34, and a face extraction circuit 36 and an average density calculation circuit 38 Is input to This average density calculation circuit 3
In step 8, the average density of one entire screen is calculated. The face extraction circuit 36 estimates the portion of the face of the person in one screen as described below, and outputs R, G, and B colorimetric data of the portion estimated as the face. The exposure calculation circuit 40 calculates the exposure using the three-color photometric data output from the face extraction circuit 36 and the average density calculated by the average density calculation circuit 38, and outputs the color correction filter 16 via the driver 42. Is controlled, and the black shutter 23 is opened and closed to perform printing. When the average density calculated by the average density calculation circuit 38 is used, an exposure correction amount for the average density can be calculated. If the exposure correction amount is not determined, the exposure amount may be determined directly from the three-color photometric data output from the face extraction circuit 36 without necessarily requiring the average density calculation circuit 38.

FIG. 2 shows a face extraction routine by the face extraction circuit 36. In step 100, noise removal, that is, smoothing of the three-color photometric data input at step 100 is performed. In the next step 102, the R, G, and B colorimetric data are converted to H (hue value) and L by the following equations (1) to (3).
(Lightness value) and S (saturation value).

L = (R + G + B) / 3 (1) S = 1-min (r ′, g ′, b ′) (2) H = H ′ / 2Pi (.) 3) However, R, G, and B are standardized such that the minimum value is 0 and the maximum value is 1 as shown in the three-dimensional color coordinates of FIG.
Colorimetric data, min () is the minimum value in parentheses, r ′, g ′, b ′ are r ′ = R / L, g ′ = G / L,
b ′ = B / L. H ′ is given by the following equation (4), and Pi (i is one of R, G, and B) is P
It is.

[0021]

(Equation 1)

However,

[0023]

(Equation 2)

In step 104, for each pixel on the original image, as shown in FIG. 4 (2), a color space of a coordinate system (hereinafter, referred to as a hue value axis, a saturation value axis, and a brightness value axis). Based on the distance V obtained in the HLS color space), an integration process (details will be described later) is performed by an iterative region expansion method. In the next step 106, a region in contact with the outer edge of the original image is removed from each of the integrated regions. In the next step 108, an area that is a candidate for a person's face as a feature image is extracted from an area other than the removed area, that is, an area near the center of the original image. In the next step 110, a face area is estimated from the area extracted as a face candidate, and R, G, B three-color photometric data of the area estimated as the face is output. Then, it is determined in step 112 whether or not printing of all frames has been completed. When it is determined that printing has been completed, this routine is terminated.

Next, the details of steps 104 to 110 will be described. FIG. 5 shows the details of step 104. In step 120, one pixel Ji (i = 1 to n: the total number of pixels) is selected from the original image by raster scanning or the like. For example, the pixel J ₁ shown in FIG.
Select In the next step 121, it is determined whether or not the selected pixel is already included in a region to which a label described later is added. In the case of a positive determination, step 122
It is determined whether or not the selection has been completed for all the pixels of the original image, and the process is repeatedly executed until the selection of all the pixels is completed.

If the selected pixel Ji is not included in the labeled area, the process proceeds to step 123,
One pixel around the selected pixel Ji (so-called 1 in the vicinity of 8)
Pixel). For example, the pixel J shown in FIG.
Select ₂ . Step 1 determines whether or not the selected pixel is already included in any of the labeled regions.
The determination is made in step 24, and in the case of a negative determination, step 12
Go to 5. In step 125, the distance V in the HLS color space is determined for the two selected pixels J ₁ and J _{2 based} on the following equation (5). That is, FIG.
As shown in (2), the pixel J ₁ in the HLS color space
Determining the distance V between the point Z ₂ corresponding to Z ₁ and pixel J ₂ point corresponding to. This distance V represents the similarity of the brightness, saturation and hue of the _two pixels J ₁ and J ₂ . That is, the similarity is low when the distance V is long, and the similarity is high when the distance V is short.

[0027]

(Equation 3)

Where dH is the difference between the hue values of the pixels J ₁ and J ₂ .

DL: Difference between brightness values of pixels J ₁ and J ₂ . dS: difference between saturation values of pixels J ₁ and J ₂ .

In the next step 126, it is determined whether or not the obtained distance V is less than a predetermined value θ, thereby determining whether or not the two pixels are similar. In step (128), a process (labeling) of giving the same label to two pixels is performed.
Proceed to. On the other hand, if they are not similar, the process proceeds to step 128 without labeling.

In step 128, it is determined whether or not the above processing has been performed on all the eight neighboring pixels. If there are eight neighboring unprocessed pixels, the process returns to step 123. When all the pixels in the vicinity of 8 have been completed, the process proceeds to step 129, where one pixel on the outermost periphery of the region (in the original image) to which the same label is added is selected. Subsequently, it is determined in step 130 whether or not the selection of all the outermost pixels has been completed. If the determination is negative, the process returns to step 123 and the above processing is executed again. If the determination is affirmative, the process returns to step 120,
The next pixel Ji is selected, and the above processing is repeatedly executed. In this way, by sequentially executing the same label assignment on the peripheral pixels having high similarity from the outermost peripheral pixel of the area assigned with the same label, pixels having similar hue, lightness and saturation on the original image can be obtained. Integration processing can be performed.

When the above integration process is completed for the screen of the original image, an affirmative determination is made in step 122, and step 13
Proceed to 1. In step 131, the average value of each of the hue value, the saturation value, and the brightness value is obtained for the pixel group to which the same label is assigned, and the obtained average value is converted to the hue value, the saturation value, and the brightness value of each pixel. replace. Next Step 13
In step 2, the predetermined value θ used in step 126 is increased by the increment value θk, and the process proceeds to step 133. The increment value θk corresponds to the degree of area expansion in the integration processing, and is set to a predetermined value (for example, 1). In step 133, steps 131 and 132
Is determined by determining the number of times (for example, four times) the execution of the integration process is performed. If the number of repetitions is less than the predetermined number, all labels assigned in step 134 are released, and the process returns to step 120. When the number of repetitions reaches the predetermined number, this routine ends. In the example of FIG. 4A, after the end of this routine, labels A to G are added as shown in FIG. 12A.

As described above, in step 104, since the adjacent pixels of the original image are integrated by the similarity of hue, lightness, and saturation, the original image is classified for each region that is more visible. Can be.

FIG. 6 shows the details of step 106 in FIG. 2. In step 136, one area on the original image of the pixel group labeled as described above is selected.

Here, as can be understood from the fact that the focusing position of the photographing device such as a camera is substantially at the center, most of the characteristic images such as the face of the person in the original image are located at the central portion of the original image. Exists. Therefore, in most cases, the region of the image that is in contact with the outer edge (image frame) of the original image is not a characteristic image such as a background. For this reason, in the next step 137, it is determined whether or not the pixel included in the selected area is located at the outermost periphery of the original image to determine whether or not the selected area is in contact with the outer edge of the original image. Judge. If it is determined that the selected area is in contact with the outer edge of the original image, the processing proceeds to step 139 after removing the selected area in step 138. In this step 138, by removing the label of only the selected area, it can be treated equivalent to the removal of the area. On the other hand, if it is determined that they are not in contact, the process proceeds directly to step 139.

In the next step 139, it is determined whether or not the above processing has been completed for all areas of the labeled pixel group in the original image, and if there is an unprocessed area, the process returns to step 136 and repeats. Execute. If the above processing has been completed for all areas, this routine ends. Therefore, after the end of this routine, a feature image, for example, a region including a face remains with high accuracy. FIG.
In the example of (1), after the end of this routine, as shown in FIG. 12 (2), the labels C and D remain.

FIG. 7 shows details of step 108 in FIG. 2. In step 140, the original image is divided for each area of the original image labeled as described above. Also,
Number each divided area. In this case, the area in contact with the outer edge of the original image has been removed.

In the next step 142, the small area is removed by judging the area of the divided area, and the numbering is performed again. In the next step 144, a large area is obtained by performing a contraction process of deleting all the boundary pixels of the region and removing it by one skin, and a dilation process of expanding the boundary pixels in the direction of the background pixel and increasing the thickness of one skin in contrast to the contraction process. Are separated from the large area. In the next step 146, the small area is removed and renumbering is performed in the same manner as in step 142, and in step 148, the same contraction and expansion processing as described above is performed to separate the weakly coupled areas from each other. In step 150, the small area is removed and renumbered in the same manner as described above. In the example of FIG.
After the end of this routine, numbers 1 and 2 are numbered as shown in FIG.

FIG. 8 shows the details of step 110. In step 162, one of the areas extracted in step 108, that is, the routine extracted in FIG. 7, is selected as the attention area, and the horizontal fillet diameter of the attention area is selected. In addition to normalizing the size of the attention area by performing scaling processing of the attention area so that the vertical fillet diameter becomes a predetermined value,
The density value or the luminance value is normalized according to the following equation (6).

[0040]

(Equation 4)

Where dmax: maximum density value (or luminance value) in the area dmin: minimum density value (or luminance value) in the area ds: full scale density value (or luminance value) of the image sensor d: density value before standardization (Or luminance value) dr: density value (or luminance value) after standardization

In step 164, a plurality of (ten in this embodiment) standard face images stored in advance (face images viewed from the front, face images viewed from the side (left and right), downward face images, upward face images, The correlation coefficient r of the attention area with respect to the face image or the like is calculated by the following equation (7), and this correlation coefficient is used as a feature amount. Even if this standard face image is data of only the face outline, the face outline data is added to the face internal structure (eye, nose,
Data).

[0043]

(Equation 5)

However,

[0045]

(Equation 6)

Where T is the length of the horizontal and vertical fillet diameters of the image (here, the lengths of the fillet diameters are the same), f
(X, y) represents a region of interest, and g (x, y) represents a standard face image.

Then, in step 166, it is determined whether or not the region of interest is a person's face by linear discriminant analysis using the above feature amount as a variable, and the R, G, B photometric data of the region determined to be a face is determined. Is output to the appropriate exposure amount calculation circuit 40. In step 168, it is determined whether or not the determination as to whether or not the face is a face has been completed for all the extracted areas. If not, steps 162 to 168 are repeated.

In the above description, the correlation coefficient is used as the feature value used to determine whether or not the face is a person. However, the invariant derived from the normalized central moment around the center of gravity, the autocorrelation function, or Geometric invariants may be used.

The central moment μ _pq around the (p + q) th order center of gravity of the image f (x, y) is _expressed by

[0050]

(Equation 7)

However,

[0052]

(Equation 8)

Then, the normalized central moment around the center of gravity is as follows.

[0054]

(Equation 9)

Here, y = (p + q + 2) / 2 p + q = 2, 3,... From the above, the following seven invariants ψ _{i are} obtained from the normalized central moments around the second and third centroids. _, (I
= 1, 2,..., 7) are derived.

[0056]

(Equation 10)

The autocorrelation function R _f is expressed as follows.

[0058]

[Equation 11]

The geometric invariant feature is expressed by the following equation.

[0060]

(Equation 12)

By doing so, the background region in contact with the outer edge of the original image is removed, and the judgment is made using the outline and internal structure of the region. Face data can be extracted with high accuracy from an image in which trees and the like are mixed.

It should be noted that the determination as to whether or not the face is a person's face may be added to the determination as to whether or not the area is within a predetermined skin color range. That is, one region is selected as a region of interest from the regions extracted in the routine of FIG. 7, and the average value of each of the hue value, lightness value, and saturation value of the region of interest is obtained, and the obtained hue value, lightness value It is determined whether or not each of the average values of the values and the saturation values is included in a predetermined range indicating the skin color. If it is within the predetermined range indicating the skin color, it is estimated that the attention area is a person's face. In this case, it may be determined whether or not there is a pixel included in the predetermined range without calculating the average value of each of the hue value, lightness value, and saturation value of the attention area.

The proper exposure amount calculating circuit 40 includes the face extracting circuit 3
6, the R, G, and B photometric data of the face area extracted as described above and the screen average density D _i (i = R, G, or B) of one frame calculated by the average density calculation circuit 38 ) and using the calculated the proper exposure amount E _i according to the following equation, and outputs to the driver 42. The driver 42 controls the light adjustment filter 16 calculates an exposure control value from the appropriate exposure amount E _i.

[0064] l _og E _i = However _{LM i · CS i · (DN} i -D i) + PB i + LB i + MB i + NB i + K 1 + K 2 ... (8), each symbol is a thing of the next.

LM: Magnification slope coefficient, which is set in advance according to the enlargement magnification determined by the type of negative and the print size.

CS: There are color slope coefficients prepared for each type of negative for underexposure and overexposure, and it is determined whether the average density of a frame to be printed is under or over the standard negative density value. Under exposure or over exposure is selected.

DN: standard negative density value. D: Average density value of print frame.

PB: Correction balance value for standard color paper, which is determined according to the type of color paper.

LB: For a standard printing lens. The correction lens balance value is determined according to the type of the printing lens.

MB: a correction value (master balance value) for a change in the printing light source and a change in the paper developing performance.

NB: Negative balance (color balance) value determined by the characteristics of the negative film.

K ₂ : color correction amount. K ₁ : a density correction amount represented by the following equation.

[0073]

(Equation 13)

[0074] Here, K _a, K _b are constants, FD is the face region average density. Further, as the correction value determined density correction amount K ₁ in equation (8) by the film detecting device may represent a color correction amount K ₂ using the face region average density as follows.

[0075]

[Equation 14]

Here, K _c is a constant. Furthermore, the density correction amount K ₁ in equation (8), the color correction amount K ₂ is a correction amount determined by the film detecting device, the average concentration FD of average density D _i a face region of the print frame in formula (8) _i
The exposure amount may be determined instead.

In this embodiment, since it is determined whether or not the extracted area after removing the background area in contact with the outer edge of the original image is a person's face, the face, ground,
Face data can be extracted with high accuracy from an image in which trees and the like are mixed.

Note that, for each of the candidate regions extracted in step 108, the shape and color information of the region of interest and the shape and color information of the neighboring region located around the region of interest determine whether the region of interest is a face. Can be determined. That is, it has the same hue value and saturation value or approximate hue value and saturation value as the attention area around the attention area, and has a size (for example, a horizontal fillet diameter and a vertical fillet diameter can be adopted). Is the size of the attention area
It may be determined whether or not an area corresponding to a person's hand or foot has been extracted by determining whether or not an area in a range of １００100% has been extracted.

Further, it is also possible to determine whether or not the attention area is a face based on the shape of the neighborhood area located around the attention area and the shape of the attention area by converting the extracted area into a line graphic. In this case, each area is converted into a line figure by performing the line information extraction processing of the area extracted as described above. Here, it is determined whether there is a line figure representing a shoulder by comparing a standard line figure representing the shoulder of a person stored in advance with a line figure on one screen, and a line figure representing a shoulder exists. In this case, if a line figure exists above the line figure, the line figure is regarded as a line figure of interest, and a line figure representing a head (eg, hat, hair, helmet, etc.) exists above the line figure of interest. to decide. Since a line figure representing the head exists above the line figure of interest and a line figure representing the shoulder exists below the line figure of interest, the line figure of interest is likely to be a line figure of the face. For this reason, it is determined whether or not the contour of the line graphic of interest is similar to the contour of the line graphic of the standard face. It is determined that the line graphic of interest is a face, and R, G, and B photometric data of an area corresponding to the line graphic of interest is output. As described above, by determining whether or not the face is a face based on the shape of the attention area or the like, face data can be extracted from an image in which faces, grounds, trees, and the like having similar hues are mixed. Further, since the face is not determined using the fine structure of the face, it is possible to determine whether the face is a face in a short calculation time even if the resolution of the determination target image is low.

FIG. 9 shows a modification in which the present invention is applied to an exposure amount determining apparatus separate from a printer or a printer processor. In FIG. 9, parts corresponding to those in FIG. Although the average density calculation circuit 38 is not always necessary, an integrated transmission density detection circuit for detecting LATD of the entire screen may be used instead.

FIG. 10 shows a face extraction circuit of FIG.
Output circuit 36₁, 36_Two... Composed of 36n, parallel processing
Is used to calculate the amount of exposure. Face extraction circuit 3
6₁, 36_Two... 36n is in the time chart of FIG.
Therefore, read the image, calculate the amount of exposure, and output the result
I do. In FIG. 11, t₁Is the image reading time for one frame,
t _TwoIs the exposure calculation time for one frame, t_ThreeIs the exposure per frame
Calculation result transfer time, t_Two>> t₁, T_ThreeIt is.
Face extraction circuit 36₁Is t₁Read one frame of image in time,
t_TwoCalculate the exposure amount with time, t_ThreeTransfer calculation results by time
I do. Face extraction circuit 36₁Reading of one frame by
At the same time, the film is sent for one frame and the face extraction circuit 3
6_TwoReading of a single frame image is started by the face extraction circuit
36₁Exposure calculation and face extraction circuit 36_TwoImage reading and
Are performed in parallel, and thereafter, the face extraction circuit 36_Three, 36
_Four.. 36n are processed in parallel.

The time Tp required for parallel processing of mxn frames is Tp = m (t ₁ + t ₂ + t ₃ ) + (n−1) t ₁ . On the other hand, the processing time Ts when the parallel processing is not performed
Is Ts = mn (t ₁ + t ₂ + t ₃ ). Therefore,

[0083]

(Equation 15)

It is possible to double the speed. This parallel processing device can be applied to the printer shown in FIG.

The present invention, besides determining the exposure amount of the photographic printing apparatus, determines the exposure amount of the digital color printer, the copy condition of the copying machine, the exposure amount of the camera, the display condition of the CRT screen, and the hard copy from the magnetic image data. Can also be applied to the determination of the amount of light when creating.

In the above embodiment, the similarity of each pixel is determined based on the distance in the color space based on the hue value, the saturation value, and the lightness value. However, the present invention is not limited to this color space. The similarity of the values, the similarity of the hue value and the saturation value may be determined.

For example, to obtain the similarity of the hue values, a histogram for the hue values of the original color image is obtained, and the obtained histogram is divided into hills or valleys of the histogram. The hue value range of each mountain is determined. Next, it is determined which hue value range the hue value of each pixel belongs to, so that it is determined which of the crests each pixel belongs to. Cluster). Subsequently, the color original image is divided into regions corresponding to the divided groups. As a result, the color original image is divided into regions each including a pixel having a hue value within the hue value range divided by the histogram. Therefore, one area on the color original image includes pixels whose hue value is within a predetermined range, and the image features are extracted from the area other than the area in contact with the outer edge of the color original image as described above. By selecting at least one region to be represented, it is possible to select a region in which data of the selected region includes data of a characteristic image such as a human face.

To determine the similarity between the hue value and the saturation value, a two-dimensional histogram of the hue value and the saturation value is obtained.
The color original image is divided in the same manner as described above by dividing the two-dimensional histogram for each mountain, and at least one region other than the region in contact with the outer edge of the color original image is selected from the divided regions to obtain data. Extract. The hue of a person's face is similar to the skin color of the ground, trees, etc., but in most cases the saturation is different and it often touches the outer edge of the original image. If data is extracted from the color region excluding the region having the same or similar hue value and saturation value, data of a characteristic image such as a human face can be extracted.

[0089]

As described above, according to the present invention, data is extracted from a color area other than a color area in contact with the outer edge of the original image based on the hue value. Can be extracted with high accuracy.

Further, since data is extracted from the color area other than the color area in contact with the outer edge of the original image based on the hue value and the saturation value, even if parts having the same or similar hue value coexist. An effect is obtained that data of the characteristic image of the original image can be extracted with high accuracy.

[Brief description of the drawings]

FIG. 1 is a schematic diagram illustrating a printer according to a first embodiment of the present invention.

FIG. 2 is a flowchart showing a face extraction routine of a face extraction circuit.

FIG. 3 is a diagram showing color coordinates.

FIG. 4A is a diagram showing an original image. (2) is a diagram showing a color space having lightness value, hue value, and saturation value as axes.

FIG. 5 is a flowchart showing details of step 104 in FIG. 2;

FIG. 6 is a flowchart showing details of step 106 in FIG. 2;

FIG. 7 is a flowchart showing details of step 108 in FIG. 2;

FIG. 8 is a flowchart showing details of step 110 in FIG. 2;

FIG. 9 is a schematic diagram of an exposure calculation device to which the present invention is applied.

FIG. 10 is a schematic diagram of an exposure calculation device that performs parallel processing by a plurality of face extraction circuits.

FIG. 11 is a diagram showing a time chart of parallel processing.

FIG. 12 is an image diagram for describing up to extraction of a face candidate region of an original image.

[Explanation of symbols]

 28 CCD image sensor 30 Amplifier 36 Face extraction circuit

Claims (3)

(57) [Claims] 1. A color original image is divided into a large number of pixels, each pixel is separated into three colors of red light, green light and blue light, and photometry is performed. Based on data obtained by the photometry, the color original image is converted into a hue. A division selected by selecting at least one color region of a color region other than a color region that is divided into color regions that are the same or similar and that is in contact with the outer edge of the color original image in the divided color regions A method for extracting characteristic image data, in which data of a region is extracted as characteristic image data. 2. A color original image is divided into a large number of pixels, each pixel is decomposed into three colors of red light, green light and blue light, and photometry is performed. The color original image is converted into a hue based on data obtained by photometry. Dividing at least one color region having the same or similar value and saturation value, and selecting at least one color region other than the color region that is the divided color region and is in contact with the outer edge of the color original image. A feature image data extraction method for extracting data of a selected divided region as feature image data. 3. The method for extracting characteristic image data according to claim 1, wherein when selecting an area, it is determined whether or not the divided area is a human face, and the area determined to be a human face is selected. .