JP2004334836A - Method of extracting image feature, image feature extracting program, imaging device, and image processing device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To extract features, such as a human face in an image, at a high speed as well as at high precision. <P>SOLUTION: In an image feature extracting method wherein images of required size are segmented, one by one from an image to be processed, each segmented image is compared with collation data of a feature image, and whether there is the feature image in the image to be processed is detected, the range of the size of the feature image, in contrast to the size of the image to be processed, is limited based on the information of distance from an object, when the image to be processed is photographed (Step S6). Thus, comparison of too large or too small segmented images, as compared to the feature image for which the collation data can be skipped, and processing at higher speed as well as with higher precision can be attained. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to an image feature extraction method, a feature extraction program, an imaging apparatus, and an image processing apparatus that can extract at high speed whether or not a feature part image such as a face exists in a processing target image.

  For example, as described in Patent Document 1 below, a digital camera is equipped with an autofocus device that extracts a face part of a subject from the screen and automatically focuses the digital camera on the extracted face. There is something that is. However, this Patent Document 1 only discloses a focusing technique, and there is no consideration as to how a facial image can be extracted at high speed by extracting the face portion of the subject.

  When extracting a face part from the screen, template matching is often used. This is because the similarity between each cut image sequentially extracted from the subject image in the search window and the face template is sequentially determined, and the face of the subject is located at the search window position that matches the face template with a similarity equal to or greater than a threshold value. It is determined that there is.

  Conventionally, when performing this template matching process, it is not known how big the subject's face is on the screen, so the size of the face template can vary from small to full screen. Many different templates are prepared, template matching processing is performed using all templates, and face images are extracted.

Japanese Patent Laid-Open No. 2001-215403

  If features such as the subject's face can be extracted before shooting, etc., the time to auto-focus on the subject's face can be reduced, and white balance photography can be performed to match the skin color of the face, etc. There are many advantages. However, there is a problem that it takes a long time to extract a face because it has been necessary to prepare a large number of face templates from small to large and perform matching using each template as before. . In addition, if a large number of template images are prepared in the memory, there is a problem that the memory capacity increases and the cost of the camera increases.

  The above-mentioned example is a case where a person is photographed with a camera. For example, an image processing apparatus or printer reads a processing target image from the camera, detects whether or not a human face is present in the processing target image, and flesh color It is convenient if a feature part image such as a face can be extracted at a high speed even in image correction adapted to the above, or for correcting red eyes by flash emission, for example.

  An object of the present invention is to provide an image feature portion extraction method, a feature portion extraction program, an imaging device, and an image processing device capable of extracting a feature portion image such as a face from a processing target image at high speed and with high accuracy. It is in.

  In the image feature portion extraction method of the present invention, an image having a required size is sequentially cut out from a processing target image, and each feature image is compared with each extracted image and comparison data of the feature portion image. An image feature portion extraction method for detecting whether or not an image exists, wherein a range of the size of the feature portion image with respect to the size of the processing target image is measured up to a subject when the processing target image is captured And limiting the size of the cut-out image to be compared with the collation data.

  With this configuration, it is possible to omit useless processing of cutting out a partial image that is too large or too small compared to the size of the characteristic partial image from the processing target image and comparing it with the collation data, thereby shortening the processing time. It becomes possible. In addition, since the size of the collation data to be used and the cut-out image is limited based on the distance information, a feature part (for example, a face) may be erroneously detected as a feature part even if the size is misregistered. Can be prevented.

  The limitation of the characteristic portion extraction method of the present invention is performed using focal length information of the photographing lens in addition to the distance information to the subject.

  With this configuration, it is possible to limit the range in which a characteristic portion (for example, a face) enters with higher accuracy.

  The comparison of the feature portion extraction method of the present invention is performed using a resized image obtained by resizing the processing target image.

  With this configuration, for example, it becomes easy to extract face images that differ depending on the individual regardless of the difference between the individuals.

  The comparison of the feature portion extraction method of the present invention is performed by changing the size of the resized image using the collation data corresponding to the feature portion image having a fixed size, or conversely, It is characterized by using the collation data in which the size is fixed and the size of the feature portion image is changed.

  With this configuration, it is possible to extract feature portion images at high speed.

  In the feature portion extraction method of the present invention, the collation data is template image data of the feature portion image.

  When extracting a feature part image such as a face image using template image data, it is preferable to prepare a plurality of types of template image data. For example, in addition to a normal person's face template, a face image can be extracted with high accuracy by preparing a template for a person wearing glasses, an elderly face template, an infant face template, and the like.

  In the feature portion extraction method of the present invention, the collation data is data obtained by quantifying the feature amount of the feature portion image.

  The digitized collation data is, for example, data obtained by digitizing pixel values (density values) for each pixel position of the characteristic partial image. Alternatively, it is collation data obtained by causing a computer to learn a face image using a machine learning algorithm such as a neural network or a genetic algorithm for an actual image. In this case as well, as with the template image, in addition to normal human face verification data, various types of data such as verification data for a person wearing glasses, verification data for an elderly face, verification data for an infant's face are prepared. It is preferable to keep it. Since the data is digitized, the memory capacity does not increase even if many types are prepared.

  The collation data of the feature portion extraction method of the present invention is characterized by comprising data describing rules for extracting feature amounts of the feature portion images.

  This configuration is the same as the digitized data, and the search range of the feature partial image in the processing target image is limited, so that the feature partial image can be extracted at high speed.

  The feature portion extraction program of the present invention limits the size range of the feature portion image with respect to the size of the processing target image based on distance information to a subject when the processing target image is captured. A step of limiting the size of the image is provided.

  By making the image feature extraction method into a program, it is possible to cause the computer-equipped device to execute this program, and various uses are possible.

  The image processing apparatus of the present invention includes the above-described feature portion extraction program, and the distance information used when the feature portion extraction program executes the step is added as tag information to the processing target image. It is distance information.

  With this configuration, the image processing apparatus can perform various correction processes. For example, brightness correction, color correction, contour correction, gradation correction, defect correction, and the like can be performed. These correction processes are not necessarily applied to the entire image, but also include correction processes for local regions in the image. If the distance information is added to the processing target image as tag information, the image processing apparatus can easily calculate the size of the feature part image in the processing target image. The search range can be narrowed.

  An imaging apparatus according to the present invention includes the above-described feature portion extraction program and means for obtaining the distance information necessary when executing the step of the feature portion extraction program.

  With this configuration, the imaging apparatus can output image data corrected to focus on a characteristic portion, for example, a human face, or to have a beautiful skin color at the time of shooting.

  The means of the imaging apparatus of the present invention includes a distance measuring sensor, a means for counting the number of motor driving pulses when the photographing lens is focused on the subject, a means for obtaining focal length information of the photographing lens, a person photographing mode, a landscape photographing mode, It includes any one of means for estimating the distance to the subject from the photographing mode such as the macro photographing mode, and means for estimating the distance to the subject from the focal length of the photographing lens.

  Since distance information can be obtained by using a distance measuring sensor or a focusing lens that is normally mounted on the imaging apparatus, the cost of the imaging apparatus can be reduced. Even if these distance measuring sensors and pulse counting means are not installed, the approximate distance to the subject can be estimated from the photographing mode and the focal length information of the photographing lens. The size of the feature portion to be extracted can be limited.

  According to the present invention, since the size of the cut-out image used in the comparison process with the collation data is limited to the size range of the feature portion image, the number of comparison processes is reduced, and the processing speed and accuracy are increased. Can be achieved.

  Hereinafter, an embodiment of the present invention will be described with reference to the drawings. In the following description, an image feature extraction method executed by a feature extraction program installed in a digital camera, which is a kind of image pickup apparatus, will be described as an example. However, image processing apparatuses including printers and image pickup apparatuses generally have similar feature portions. A similar effect can be obtained by installing an extraction program.

(First embodiment)
FIG. 1 is a configuration diagram of a digital still camera according to the first embodiment of the present invention. This digital still camera includes a solid-state imaging device 1 such as a CCD or a CMOS, a lens 2 and a diaphragm 3 placed in front of the solid-state imaging device 1, and correlated double sampling with respect to an image signal output from the solid-state imaging device 1. An analog signal processing unit 4 that performs processing, an A / D conversion unit 5 that converts an analog signal-processed image signal into a digital signal, gamma correction, synchronization processing, and the like for the image signal converted into a digital signal A digital signal processing unit 6 to be applied, an image memory 7 for storing an image signal processed by the digital signal processing unit 6, and an image signal (shooting data) stored in the image memory 7 when the user presses the shutter button. Is recorded on an external memory or the like, and a display unit 9 is provided on the back of the camera or the like and displays the stored contents of the image memory 7 through.

  The digital still camera further includes a control circuit 10 including a CPU, a ROM, and a RAM, an operation unit 11 that receives an instruction input from a user and performs an on-demand display process on the display unit 9, and an output from the image sensor 1. Then, the image signal processed by the digital signal processing unit 6 is taken in. Based on an instruction from the control circuit 10, as will be described in detail later, a feature extraction portion of the subject, in this example, a face extraction processing portion 12 that extracts a face portion, and the lens 2 A lens driving unit 13 that performs focusing and magnification control based on an instruction signal from the control circuit 10, a diaphragm driving unit 14 that controls an aperture amount of the diaphragm 3 based on an instruction signal from the control circuit 10, and solid-state imaging An image sensor control unit 15 that controls the driving of the element 1 based on an instruction signal from the control circuit 10 and a subject based on the instruction signal from the control circuit 10. And a distance measuring sensor 16 to measure the distance.

  FIG. 2 is a flowchart showing a processing procedure of face extraction processing performed by the face extraction processing unit 12 in accordance with the face extraction program. The face extraction program is stored in the ROM of the control circuit 10 shown in FIG. 1, and the face extraction processing unit 12 functions when the CPU reads the face extraction program into the RAM and executes it.

  The imaging device 1 of the digital still camera always outputs an image signal at a predetermined cycle even before the user presses the shutter button, and the digital signal processing unit 6 performs digital signal processing on each image signal. The face extraction processing unit 12 sequentially captures this image signal and performs the following processing on each input image.

  First, the size of the input image (processing target image) is acquired (step S1). For example, in the case of a camera in which the size of the input image used for the face extraction process is different depending on whether the user is shooting with 640 × 480 pixels or 1280 × 960 pixels, this size information is acquired. . When the size of the input image is fixed, this step S1 is not necessary. Next, distance information to the subject measured by the distance measuring sensor 16 is acquired from the control circuit 10 (step S2).

  Even if the camera does not include the distance measuring sensor 16, if the camera has a mechanism for driving the focusing lens back and forth to focus on the subject, the number of pulses driven by the motor is counted. Distance information can be obtained from numerical values. In this case, the relationship between the pulse count value and the distance may be held as a function or as table data.

  In the next step S3, it is determined whether or not the zoom lens is used. If the zoom lens is used, the zoom position information is obtained from the control circuit 10 (step S4), and then the focus of the lens is obtained. Distance information is acquired from the control circuit 10 (step S5). If it is determined in step S3 that the zoom lens is not used, the process skips step S4 and proceeds to step S5.

  Based on the input image size information and the lens focal length information acquired by the above processing steps, it is possible to determine the size of the face of the person who is the subject in the input image. Therefore, in the next step S6, the upper limit and lower limit ranges of the search window size according to the face size are determined.

  As shown in FIG. 3, the search window is a window 23 having the same size as the face image with respect to the processed image 21 to be subjected to the template matching process, that is, the size of the template 22 shown in FIG. The normalized cross-correlation coefficient between the image cut out by the search window 23 and the template 22 is obtained by the following processing steps, the matching degree is calculated, and when the matching degree, that is, the similarity does not reach the threshold value, the search window 23 is shifted in the scanning direction 24 by a certain amount of pixels on the processed image 21, for example, one pixel, and an image for the next matching processing is cut out.

  Here, the processed image 21 is an image obtained by resizing the input image. For example, rather than performing a matching process using a high-definition input image of 1280 × 960 pixels as a processed image, an image obtained by resizing this input image to, for example, 200 × 150 pixels is used as a processed image, and a template (of course, a face on the template side) The image is not a high-definition face image, but a face image having a small number of pixels, for example, 20 × 20 pixels is used.) It is easier to detect a general “face” by ignoring individual differences. Become.

  In the next step S7, it is determined whether or not the size of the search window is within the range, that is, whether or not it is within the upper and lower limits of the face size in the processed image 21. Next, it is determined whether or not there is a template 22 having a size that matches the size of the search window 23 (step S8). If it exists, the corresponding template is selected (step S9). If it does not exist, the template is resized to generate a template that matches the size of the search window 23 (step S10), and the process proceeds to the next step 11. .

  In step S11, template matching processing is performed while scanning the search window 23 along the scanning direction 24, and it is determined whether or not there is an image portion having a similarity equal to or greater than the threshold value α.

  If there is no image portion whose similarity is equal to or greater than the threshold value α, the process proceeds to step S12, the size of the search window 23 is changed as shown in FIG. 5, and then the size of the search window 23 to be used is changed. After the determination, the process proceeds to step S7. Thereafter, step S7 →... → step S11 → step S12 → step S7 is repeated.

  Thus, in the present embodiment, the size of the template is changed as shown in FIG. 6 while changing the size of the search window 23 from the upper limit value to the lower limit value (or from the lower limit value to the upper limit value) as shown in FIG. Also change the template matching process.

  If an image portion showing a similarity equal to or greater than the threshold value α is detected in step S11, the process proceeds to face detection determination processing in step S13, the face position is specified, and the position information is output to the control circuit 10 to output this face. The detection process ends.

  If step S7 →... → step S11 → step S12 → step S7 is repeated and the size of the search window 23 falls outside the upper and lower limits, the determination result in step S7 is negative (N). . In this case, the process proceeds to the face detection determination process in step S13, and “no face” is determined.

  In this embodiment, since the processing system emphasizes the processing speed, when a portion having a similarity of α or more in the input image (processing target image) is detected in step S11, that is, one person When a face image is extracted, the process immediately proceeds to step S13 to end the face image search process.

  However, for example, in the case of a processing system emphasizing the detection accuracy of face images, matching processing is performed on all clipped images and all templates to obtain respective similarities, and finally the highest similarity is shown. An image part is detected as a face image, or all image parts having a similarity equal to or higher than a predetermined value are detected as face images. This is not limited to the first embodiment, and the same applies to the second, third, fourth, and fifth embodiments described later.

  In the first embodiment, the face image search process is performed using one type of template shown in FIG. 4. However, a plurality of types of templates are prepared, and a face image is detected using each type of template. It is good to have a configuration to do. For example, in addition to an ordinary person's face template, a face template with glasses, an elderly face template, an infant face template, etc. are prepared so that highly accurate face image extraction processing can be performed. It becomes possible.

  As described above, according to the present embodiment, a plurality of types of templates used in the template matching process are prepared, and the matching process using each template is performed. The upper and lower limits of the template to be used based on the distance information to the subject. Therefore, the number of template matching processes can be reduced, and the face extraction process can be performed with high accuracy and high speed.

(Second Embodiment)
FIG. 7 is a flowchart showing the processing procedure of the face extraction program according to the second embodiment of the present invention. The configuration of a digital still camera equipped with this face extraction processing program is the same as that shown in FIG.

  In the first embodiment described above, the template matching process is performed while changing the size of the search window and the template. However, in this embodiment, the sizes of the search window and the template are fixed and the size of the processed image 21 is changed. Perform template matching while resizing the.

  Steps S1 to S5 are the same as those in the first embodiment. Following step S5, in the present embodiment, the upper and lower limits of the size of the processed image 21 are determined (step S16). In the next step S17, it is determined whether or not the size of the processed image 21 is within a range corresponding to the upper and lower limits of the face size.

  If it is determined in step S17 that the size of the processed image 21 is within the upper and lower limits, the process proceeds to step S11, where template matching processing is performed, and an image whose similarity is greater than or equal to the threshold value α. Determine if the part exists. If an image part having a similarity equal to or greater than the threshold value α cannot be detected, the process returns from step S11 to step S18 to resize the processed image 21, and the template matching process is repeated. If an image portion equal to or greater than the threshold value α is detected, the process proceeds from step S11 to the face detection determination process in step S13, the face position is specified, the position information is output to the control circuit 10, and the face detection process is terminated. .

  After the size of the processed image changes from the upper limit value to the lower limit value (or after the change from the lower limit value to the upper limit value) due to resizing of the processed image 21, the determination result of step S17 becomes negative (N). In this case, the process proceeds to step S13, and “no face” is determined.

  In this way, in this embodiment, the size of the subject's face relative to the input image is limited based on the distance information to the subject, so the number of template matching processes can be reduced, and face extraction can be performed with high accuracy and high speed. Can be done. Moreover, since only one template is prepared in advance, the storage capacity of the template can be reduced.

(Third embodiment)
FIG. 8 is an explanatory diagram of a digital still camera according to the third embodiment of the present invention. In the first and second embodiments, the distance information to the subject is acquired by the distance measuring sensor 16, but in this embodiment, the distance information to the subject is acquired without using the distance measuring sensor, and the template matching process is performed. Perform face extraction.

  For example, when taking a commemorative photo of a subject with a digital still camera installed in a studio, or when a camera installation position and a monitoring target location such as a door entrance are fixed like a surveillance camera, the subject 25 and a digital still camera The distance to 26 is known. Further, when the installation base 27 of the digital still camera 26 is moved by a moving mechanism such as a motor and a rail, the movement amount is obtained from a motor timing belt, a rotary encoder, or the like of the moving mechanism, whereby the control circuit of FIG. 10 can know the distance to the subject 25.

  Note that the digital still camera according to the present embodiment includes a mechanism that acquires position information from a moving mechanism instead of having no distance measuring sensor in the configuration of FIG. Alternatively, position information input from the operation unit 11 by the user is used.

  FIG. 9 is a flowchart showing the processing procedure of the face extraction program according to the present embodiment. In the face extraction program of this embodiment, first, distance information between reference points (the default installation position of the camera and the subject position) shown in FIG. 8 is acquired (step S20), and then the same as step S1 of the first embodiment. In addition, the size of the input image is acquired.

  In the next step S21, information on how much the moving mechanism has moved with respect to the subject 25 is acquired from the control circuit 10, and the process proceeds to step S3. Since the process of the following step S4-step S13 is the same as 1st Embodiment shown in FIG. 2, the description is abbreviate | omitted.

  As described above, also in this embodiment, the size of the face of the subject with respect to the input image is limited based on the distance information to the subject, so the number of template matching processes can be reduced, and face extraction can be performed with high accuracy and high speed. Can be done.

(Fourth embodiment)
FIG. 10 is a flowchart showing the processing procedure of the face extraction program according to the fourth embodiment of the present invention. This embodiment is also a program applied to a surveillance camera or the like as described with reference to FIG. 8. First, distance information between reference points shown in FIG. 8 is acquired (step S20), and then step S1 of the second embodiment. As with, the size of the input image is acquired.

  In the next step S21, information on how much the moving mechanism has moved with respect to the subject 25 is acquired from the control circuit 10, and the process proceeds to step S3. Since the processing of the following steps S4 to S13 is the same as that of the second embodiment shown in FIG. 7, the description thereof is omitted.

  As described above, also in this embodiment, the size of the face of the subject with respect to the input image is limited based on the distance information to the subject, so the number of template matching processes can be reduced, and face extraction can be performed with high accuracy and high speed. Can be done. Moreover, since only one template is prepared in advance, the storage capacity of the template can be reduced.

(Fifth embodiment)
In the above-described embodiment, the template image data is used as the matching data of the feature portion image. However, it is possible to perform comparison and matching with the clipped image by the search window without using the template image data.

  For example, collation data in which the density value of each pixel of the template image of FIG. 4 is digitized in association with each pixel position coordinate is prepared, and comparison collation can be performed using this collation data. Alternatively, it is possible to extract a correlation between pixel positions with high density (positions of both eyes in the example of FIG. 4) as collation data, and perform comparison collation using the collation data.

  In this embodiment, a machine learning algorithm such as a neural network or a genetic algorithm is used for an actual image captured by an imaging device, and a feature partial image, for example, a facial image feature is learned in advance by a computer. The learned result is stored in the memory of the imaging apparatus as collation data.

  FIG. 11 is a diagram illustrating a configuration example of collation data obtained as a result of prior learning. The pixel value v_i and the score p_i are determined by learning for each pixel position in the search window. The pixel value is numerical data such as a pixel density value. The score is an evaluation value.

  When the template image is used, the evaluation value is “similarity”, which is an evaluation value as a result of comparison with the entire template image. In the case of the collation data of the present embodiment, the size of the search window is set. On the other hand, an evaluation value is determined for each pixel.

  For example, when the pixel value at a certain pixel position is “45”, the score is “9”, the faceness is set to be high, and when the pixel value at another pixel position is “10”, the score is set. Is “−4”, and it is set not to look like a face.

  Then, as a result of the comparison and collation, a cumulative value of evaluation values for each pixel is obtained, and by determining whether or not the face image is based on the cumulative value, the face image can be detected. In the collation data using the numerical data, it is preferable to prepare collation data for each size of the search window and detect a face image by each collation data.

  When a certain search window is selected and collation data for the size of the search window is not prepared, a process corresponding to the process of step S10 in FIG. It is also possible to create collation data for. For example, the pixel value is determined by interpolation calculation using a plurality of pieces of collation data having a size around that of the search window.

  The template is data obtained by extracting the feature quantity from the feature partial image as an image. The digitized collation data is data obtained by extracting the feature quantity from the feature partial image as numeric data. Therefore, instead of preparing collation data as a template or digitized data, prepare collation data in which rules for extracting feature amounts from feature partial images are described as words, and an image cut out from a processing target image using a search window It is also possible to adopt a configuration for performing the comparison process. In this case, the arithmetic processing unit of the control circuit must interpret the rules one by one, but since the range of the size of the face image is limited by the distance information, high-speed processing is possible.

  In each of the above-described embodiments, the digital still camera has been described as an example. However, the present invention can also be applied to other digital cameras such as a digital camera mounted on a mobile phone or a digital video camera that performs moving image shooting. . In addition, the distance information to the subject is not limited to the case of using a measurement value or a known value of a distance measuring sensor, the distance information acquisition method may be any method, and the extraction target is not limited to the face, but other features. The present invention can also be applied to portions.

  The feature extraction program of each embodiment described above is not limited to being installed in a digital camera. For example, the feature extraction program can be installed in a photographic printer or image processing apparatus to extract a feature portion of a subject with high accuracy and high speed. Is possible. In this case, distance information and zoom information are required to limit the size of the template or the size of the processed image to the upper and lower limits of the feature partial image. Are used as tag information added to captured image data.

  In the above-described embodiment, the range of the size of the characteristic portion included in the image is determined based on the distance information to the subject obtained by the distance measuring sensor, the number of motor driving pulses when the photographing lens is focused on the subject, and the like. Although limited, the present invention can be applied if the approximate range can be limited even when the size range of the feature portion is not known with high accuracy.

  For example, the distance to the subject can be roughly limited from the focal length of the taking lens, and if the shooting mode such as the portrait shooting mode, the landscape shooting mode, or the macro shooting mode is known, the distance to the subject can be determined. The distance can be estimated, and the feature portion extraction process can be speeded up by roughly limiting the size of the feature portion.

  Furthermore, these information may be combined, for example, the approximate distance to the subject may be estimated from the shooting mode and the focal length of the shooting lens, or the shooting mode and the number of motor driving pulses may be combined for determination.

  Since the present invention can extract a feature part image such as a face from an input image at high speed, for example, brightness correction, color correction, contour correction, gradation correction, defect correction, etc. are not limited to correction of the entire image, Correction for a local region in an image can also be performed at high speed, and it is preferable to be mounted in an image processing apparatus or an imaging apparatus.

1 is a configuration diagram of a digital still camera according to a first embodiment of the present invention. It is a flowchart which shows the process sequence of the face extraction program mounted in the digital still camera shown in FIG. It is explanatory drawing of the scanning by a search window. It is a figure which shows an example of the template of a face. It is explanatory drawing of the example which changes the magnitude | size of a search window. It is explanatory drawing of the example which changes the magnitude | size of a template. It is a flowchart which shows the process sequence of the face extraction program which concerns on the 2nd Embodiment of this invention. It is a figure which shows the example of installation of the digital still camera which concerns on the 3rd Embodiment of this invention. It is a flowchart which shows the process sequence of the face extraction program which concerns on the 3rd Embodiment of this invention. It is a flowchart which shows the process sequence of the face extraction program which concerns on the 4th Embodiment of this invention. It is explanatory drawing of the collation data based on the 5th Embodiment of this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Image sensor 2 Lens 3 Aperture 6 Digital signal processing part 10 Control circuit 12 Face extraction processing part 16 Distance sensor 21 Processed image 22 Face template 23 Search window 24 Scanning direction

Claims (12)

  An image for detecting whether or not the feature portion image exists in the processing target image by sequentially cutting out images of a required size from the processing target image and comparing each cut image with the matching data of the feature portion image The feature portion extraction method according to claim 1, wherein the range of the size of the feature portion image with respect to the size of the processing target image is limited based on distance information to a subject when the processing target image is imaged. A method for extracting a feature portion of an image, wherein the size of the cut image to be compared is limited.   2. The image feature portion extraction method according to claim 1, wherein the limitation is performed using focal length information of a photographing lens in addition to distance information to the subject.   The image feature portion extraction method according to claim 1, wherein the comparison is performed using a resized image obtained by resizing the processing target image.   4. The image feature portion extraction method according to claim 3, wherein the comparison is performed by changing the size of the resized image using the matching data corresponding to a feature portion image having a predetermined size.   The image feature portion extraction method according to claim 3, wherein the comparison is performed using the collation data in which a size of the resized image is fixed and a size of the feature portion image is changed.   6. The image feature portion extraction method according to claim 1, wherein the collation data is template image data of the feature portion image.   6. The image feature portion extraction method according to claim 1, wherein the collation data is data obtained by digitizing a feature amount of the feature portion image.   6. The image feature portion extraction method according to claim 1, wherein the collation data includes data describing a rule for extracting a feature amount of the feature portion image.   9. A feature portion extraction program for executing the image feature portion extraction method according to claim 1, wherein a range of the size of the feature portion image with respect to the size of the processing target image is determined. A feature portion extraction program comprising a step of limiting based on distance information to a subject when a target image is captured and limiting the size of the cutout image.   10. The distance information added as tag information to the processing target image is used as the distance information that is mounted when the characteristic portion extraction program according to claim 9 is mounted and the characteristic portion extraction program executes the step. An image processing apparatus.   An image pickup apparatus comprising: the feature portion extraction program according to claim 9; and means for obtaining the distance information necessary when executing the step of the feature portion extraction program.   The means includes a distance measuring sensor, a means for counting the number of motor driving pulses when the photographing lens is focused on the subject, a means for obtaining focal length information of the photographing lens, a photographing mode such as a person photographing mode, a landscape photographing mode, and a macro photographing mode. 12. The image pickup apparatus according to claim 11, further comprising: means for estimating a distance from the mode to the subject, or means for estimating a distance from the focal length of the photographing lens to the subject.

Images