JP2005086516A - Imaging device, printer, image processor and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To automatically select only image data including a specific person and to classify and duplicate the image data without displaying the image data as an image on a display device. <P>SOLUTION: A face recognizing means is used to detect a face included in the image data, feature information that can determine an individual is prepared from the detected face, and the feature information is attached to the image data so as to be stored. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to processing of image data obtained by an imaging device using a photoelectric conversion element.

  2. Description of the Related Art Conventionally, there is a system (hereinafter referred to as a direct print system) that connects an imaging device such as a digital camera to a printing device such as a printer and prints image data of a captured subject. If you want to print only image data that shows a specific person in this system, you can either output the image data to a display device such as a liquid crystal monitor equipped with a digital camera or printer to display a list of images, or The user confirms each image by sequentially switching the image depending on the member and selects image data to be printed (Patent Document 1).

  Some digital cameras, personal computers (hereinafter referred to as PCs) and the like have a slide show function for sequentially switching stored image data and displaying them as images. With this function, the user selects a plurality of image data as a target by displaying a list of images or sequentially switching images with an operation member (Patent Document 2).

In addition, when RAW image data obtained by digitizing the output value of an image sensor of a digital camera is compressed and converted into data in JPEG format by an information processing device such as a PC (hereinafter referred to as development processing) Even when image data is divided into folders and classified by a digital camera or PC, the user can select the target image data by displaying a list of images or switching the images in order using an operation member. Yes.
Japanese Patent Laid-Open No. 11-282446 JP 11-331739 A

  As described above, as a case where selection of image data by the user is necessary, for example, a case where an image captured when traveling by a plurality of people is distributed or printed can be considered. Images taken when traveling with multiple people include only those who have traveled, those that are different from those who have traveled, and only those who are different from those who have traveled. It is conceivable that there are things that only exist, or that only the scenery is reflected, and those in various situations. From the image data that contains these images, print only the image data of the person, divide it into folders for each person, save it to a portable storage medium, or transfer it. When the user selects the target image data while checking the image displayed on the display device, the time that the user spends on the selection work increases if the total number of image data is large. The operation becomes complicated.

  Further, in the case of the direct print system, if a plurality of the same image data is to be printed, an operation for the user to designate the number of prints is added.

  Further, even when the same image data is displayed on the liquid surface monitor, the color of the output image may be different depending on whether it is printed on the printer. This is often not known unless you actually print.

  Also, when developing image data in RAW format with an information processing device such as a PC, if the user wants to perform image processing so that the intended person has the optimum color and brightness, the user sets the parameters used for development. There was a need to do.

  The present invention has been made in view of such problems, and by using face recognition means when signal processing image data, the detection results of faces included in images and various types of detected faces are disclosed. It is a main object to provide an imaging apparatus having a function of creating information and adding and storing it on signal-processed image data.

  It is another object of the present invention to provide an imaging device, a printing device, and an image processing device that perform signal processing of image data based on various types of information related to a face added to the image data.

  Moreover, it aims at providing the program for performing these processes.

  In order to achieve the above object, the invention described in claim 1 of the present application includes image data in an image pickup apparatus that obtains image data by converting the output of an image pickup element that converts incident light into an electrical signal into a digital value. Feature point information for identifying a person's face to be extracted, and image data is stored with the feature point information.

  Similarly, in order to achieve the above-mentioned object, the invention according to claim 2 of the present application provides a storage means in an imaging apparatus that obtains image data by converting an output of an imaging device that converts incident light into an electrical signal into a digital value. Is read out, feature point information for specifying a human face included in the image data is extracted, and feature point information is attached to the image data.

  Similarly, in order to achieve the above object, the invention according to claim 16 of the present invention stores comparison information provided for each individual in advance in a printing apparatus for printing image data transferred from the outside. Feature point information for identifying a person's face included in the transferred image data is extracted, and the image data is stored with the feature point information attached, and a plurality of image data is set for each person. Only the image data determined to contain the same person is selected using the comparison information, and the selected image data is printed.

  Similarly, in order to achieve the above object, the invention described in claim 17 of the present invention stores comparison information provided in advance for each individual in a printing apparatus for printing image data transferred from the outside. Extracts feature point information for identifying a person's face included in the transferred image data, stores the image data with the feature point information attached, and uses the feature point information to find a face area with a person's face The image data is processed and printed so that the exposure value of the face area is appropriate.

  Similarly, in order to achieve the above object, an invention according to claim 18 of the present application is an image processing apparatus that compresses image data obtained by the imaging apparatus having the above-described configuration. Image data whose attached feature point information satisfies a predetermined condition is selected and compressed.

  Similarly, in order to achieve the above object, an invention according to claim 21 of the present application is an image processing apparatus for compressing image data obtained by an imaging apparatus having a vapor structure, and includes a feature point attached to the image data. A face area where a person's face is located is determined using information, and the image data is processed and compressed so that the exposure value of the face area is appropriate.

  Similarly, in order to achieve the above-mentioned object, the invention according to claim 25 of the present application reads out the image data from the storage means and specifies the feature point information for specifying the face of the person included in the read-out image data. Is extracted, and the image data is stored with the feature point information attached thereto.

  According to the present invention, it is possible to identify an individual included in the image, determine a person's facial expression, and the like without displaying the image data as an image on a display device.

  In addition, without displaying image data as an image on a display device, only image data including a specific person is automatically selected, or a specific person's face is displayed with optimal brightness or color expression. Image data can be automatically processed.

  Further, the image data can be classified and duplicated without displaying the image data as an image on the display device.

  These actions not only simplify the operation of the user, but also lead to a reduction in time because the processing is automated.

  Embodiments of the present invention will be described below with reference to the drawings.

  FIG. 1 is a block diagram simply showing the configuration of a digital camera which is an image pickup apparatus of the present embodiment.

  Reference numeral 11 denotes an imaging lens for forming a subject image on the imaging element 13, and 12 denotes a stop. Reference numeral 13 denotes an image sensor such as a CCD that converts incident light into an electrical signal. An A / D converter 14 converts the output signal of the image sensor 13 into a digital value to obtain image data. An image signal processing unit 15 performs processing such as gamma correction on the image data output from the A / D conversion unit. A compression processing unit 16 compresses the image data output from the image signal processing unit 15. Reference numeral 17 denotes a face detection processing unit that refers to the image data output from the compression processing unit 16 to detect whether or not a person's face has moved to the image, and to create information about the detected face. The face detection processing unit 17 may be configured to be included in the image signal processing 15. A memory 18 temporarily stores the image data obtained by the compression processing unit 16 and the feature point information obtained by the face detection processing unit 17. The feature point information will be described later. Reference numeral 19 denotes a storage medium for storing feature point information and image data in the memory 18, which may be fixed to the image pickup apparatus or detachable. Reference numeral 20 denotes a drive unit that drives the image sensor 13 and the A / D conversion unit 14. Reference numeral 21 denotes an aperture control unit that controls and drives aperture blades that adjust the amount of light incident on the image sensor 13. A lens control unit 22 drives the imaging lens 22 to adjust the focal length. A release switch 23 instructs the start of imaging. Reference numeral 24 denotes a system control unit that controls operation of the entire digital camera.

  The system control unit 24 performs, for example, control according to the operation of the release switch 23, adjustment of the focal length and exposure amount, and control of writing and reading data to and from the memory 18 and the storage medium 19. The system control unit 24 reads out and executes a predetermined program, thereby executing functions in the present embodiment.

  The face detection processing unit 17 has a face detection function as described in JP-A-2002-251380, for example.

  First, extract the edges from the image data using luminance information, detect shapes that are thought to be organs that make up the face, such as both eyes and mouth, extract the surrounding contour lines that contain them, and determine the face area To do.

  From the determined face area, points that become further features (hereinafter referred to as feature points) are detected in detail using luminance information. From the feature points, more detailed shapes and coordinates such as organs constituting the face such as both eyes and nose and the outline of the face are obtained. Thereby, for example, it is possible to obtain effective information for specifying an individual such as the distance between both eyes, the shape of the nose, the mouth, and the contour, and the arrangement state thereof. If individual feature point information is stored in the digital camera in advance, it can be determined whether the image data obtained by imaging is the same person as the individual. Note that it is possible to recognize an individual more accurately by using color information (for example, the color of a pupil or skin) in addition to information on the shape and coordinates of a face region. Further, feature point information corresponding to a plurality of face orientations is stored in advance so that the face included in the image data can be dealt with even when the face is inclined.

  In addition, the detected feature point information is compared with various comparison feature point information stored in advance in the imaging device, and a specific facial expression of the subject (whether laughing or eyes are open) ) Can also be determined. If the determination accuracy increases, it is possible to determine the gender and approximate age.

  However, in order to make a detailed determination, it is necessary to prepare data as a comparison reference in advance. By accumulating detection results each time face detection of the same person is performed, it is possible to increase the accuracy of data serving as a comparison reference.

  Note that this face detection function can be applied to RAW format image data in which the output of the image sensor is simply digitized, or JPEG format image data that has been compressed by performing processing such as correction. Any image data that can determine the shape of the subject may be used.

  Next, a flowchart from when the release switch of the digital camera in this embodiment is pressed until image data is stored will be described with reference to FIG.

  When the user operates the release switch 23, the imaging device detects this and starts an imaging operation (step 101).

  The output of the image sensor that has received the light flux from the subject that has passed through the imaging lens 11 and the aperture 12 is converted into image data (RAW format image data) of a digital signal by the A / D converter 14 (step 102).

  The image signal processing unit 15 performs a blackening process, white balance gain value calculation, aperture correction, and the like on the output data of the A / D conversion unit 14 to create a luminance signal Y, and simultaneously performs color interpolation, matrix conversion, and the like. Thus, the color difference signals U and V are created to form a YUV format signal (step 103).

  The formed YUV image data is output to the compression processing unit 16 to generate JPEG image data, which is one of the image compression formats, and stores it in the memory 18 (step 104). The data may be stored in the memory 18 in the RAW format without being compressed.

  Further, the image data in the YUV format is referred to by the face detection processing unit 17, and all the faces existing in the image are detected by the face detection function described above (step 105).

  If no face is detected, information indicating that fact is stored in the memory 18 (step 111).

  If a face is detected, the process proceeds to feature point detection processing (step 106).

  The output of the area determined to be a face is sent to the feature point detection block in the face detection processing unit 17 to detect the feature points, and from the arrangement state of these feature points, for example, the shape of each organ, both eyes And the like and the shape of the nose and mouth are stored in the memory 18 (step 107).

  The digital camera stores pre-registered information on individual feature points and additional information associated therewith, and compares it with the feature points detected this time (step 108).

  As a result of the comparison, if the feature point information is approximated (or matched), additional information for each person stored in advance in the digital camera is stored in the memory 18 (step 109). This additional information is detailed information that cannot be detected from facial feature points, such as information indicating name, age, pupil color, and the like. The color of the pupil can be obtained from the image data at the time of face detection. However, since image data having a color different from that of the original pupil may be obtained due to a red eye phenomenon or the like, it is preferable to prepare it as additional information. .

  Further, the feature point information of the image data is sent to the expression recognition block in the face detection processing unit 17, and the feature point information is used to determine the expression such as whether the face is smiling or whether both eyes are open, and the orientation of the face. Judgment is made and the result is stored in the memory 18 (step 110).

  The header information of the image data converted into the JPEG format is created using the feature point information, the additional information, and the facial expression information stored in the memory 18 (step 111), and the header information is added to the image data in the JPEG format or the like. The data is attached and stored in the storage medium 19 (step 112). In addition, when a person cannot be extracted because the person is not included in the data in the image, the fact is created by using header information to that effect.

  Note that the additional information and information obtained by the facial expression recognition block are not necessarily required. If there is feature point information of image data, it is possible to identify an individual.

  In the above flowchart, the feature points are extracted from the YUV format image data. However, as described above, the feature points can also be applied to the RAW format image data obtained by digitizing the output of the image sensor.

  Here, a method of registering feature point information and additional information that are used as a criterion for individual discrimination will be described. First, image data of a person who wants to store feature point information or additional information that is a criterion for discrimination is obtained. This may be actually captured by a digital camera or read from an external storage medium.

  Next, the obtained image data is sent to the face detection processing unit 17, the face area is detected in the same manner as in step 105 described above, and the detected face area is displayed on a liquid crystal monitor or the like. Then, the user selects an image of a person to be registered while looking at the liquid crystal monitor or the like.

  When a person to be registered is selected, feature points are detected in the same manner as in steps 106 and 107 described above, and an arrangement state of the feature points is obtained. Further, the user inputs detailed information (name, age, pupil color, etc.) of the person through the user interface with respect to the selected image. At this time, information such as sex obtained from the feature points obtained by the face detection processing unit 17 is displayed so that correction can be performed if there is an error. By performing such processing, it is possible to store in advance the feature point information that is a criterion for individual discrimination and additional information associated therewith in the digital camera. By repeatedly registering the same person and accumulating feature point information, the accuracy of determining an individual is improved.

  Next, how to use the image data accompanied by the header information will be described with two examples in the direct print system.

  One is that when printing is performed by directly connecting the above digital camera and a printer, which is a printing apparatus, without using an external information processing apparatus such as a PC, digital printing is performed using header information attached to image data. Image data in which a specific person registered in advance in the camera is captured is automatically searched, previewed on a liquid crystal monitor, etc., and printed.

  Second, when printing an image of a specific person registered in advance in the digital camera, the optimum parameter for that person is automatically obtained from the information stored in the header to correct the image data. To perform printing.

  First, an operation for automatically retrieving and printing an image in which a specific person is shown will be described with reference to FIG.

  The digital camera stores a list of feature point information and additional information that serve as a reference for identifying a person by person, and the LCD monitor displays a list of persons storing the list. Is done. When the user selects a person to be printed from the list, the digital camera reads a list corresponding to the person selected by the user (step 201).

  The feature point information described in the header of the image data stored in the storage medium is compared with the feature point information stored in the list. If the feature points approximate or match, the image data is selected by the user. It is determined that the person has been photographed, and a flag indicating that the image is to be printed is set in the image (step 202). If it is determined that the selected person is not captured, the process proceeds to the next image data processing.

  If it is determined that the selected person is photographed, the image data is displayed on a display device such as a liquid crystal monitor (step 203). If the image data displayed here does not require printing, the flag can be canceled.

  Next, it is determined whether or not it is an automatic print correction mode in which appropriate image correction is automatically performed on the face area of the selected person (step 204). If the print correction mode is not set, the displayed image is printed as it is (step 209), and the process proceeds to the next image data processing.

  If the automatic print correction mode is set, the face area of the selected person is determined from the feature point information described in the header of the image data (step 205).

  Luminance correction is performed so that the luminance value of the selected face area is optimized (step 206).

  Next, color correction is performed so that the skin color of the selected face area is optimized (step 207).

  Further, moderate sharpness correction is performed on the selected face area, and processing is performed so that the selected face area stands out (step 208).

  Then, an image subjected to luminance correction, color correction, and sharpness correction is printed by a printer (step 209).

  All images stored in the storage medium or all images included in the range selected by the user (for example, all images in the folder selected by the user or all images selected according to the shooting date and time) If the comparison between the feature point information described in the header and the list in which the individual feature point information stored in advance in the digital camera has not been completed, the process proceeds to the next image data processing and ends. If so, this flow is finished (step 210).

  Hereinafter, the image correction performed from step 206 to step 208 will be briefly described.

  First, correction for setting the face area to an optimum luminance value (exposure value) will be described. This corresponds to the processing performed in step 206 described above.

  The face area of the person is obtained from the coordinates of the feature points described in the header of the image data, the luminance signal in the face area is calculated, and the average value Yave is calculated.

The brightness correction coefficient k is calculated by the following formula using the preset optimum brightness value Yref and average value Yave of the human face area.
k = Yref / Yave
If the brightness correction coefficient k is smaller than a preset lower limit value of the brightness correction coefficient, the brightness correction coefficient k is set to this lower limit value. On the contrary, if the luminance correction coefficient k is larger than the preset upper limit value of the luminance correction coefficient, the luminance correction coefficient k is set to this upper limit value. The range of the luminance correction coefficient k is set to suppress overcorrection. The lower limit value of the luminance correction coefficient k is set to such an extent that the color balance in the saturation region is not lost, and the upper limit value is set to an extent that prevents noise deterioration. It is preferable.

  The obtained luminance correction coefficient k is applied to the RGB values of the entire image data. Thereby, the brightness | luminance of the calculated | required face area can be optimized. At this time, it is desirable to change the upper limit value of the luminance correction coefficient in accordance with the area (size) of the selected face area as shown in FIG. This is to reduce the degree of correction when the face area is less than a certain value, and to prevent the brightness balance of the entire image from changing greatly just to adjust the brightness of the face that is only small. .

  If there are two or more persons registered in advance in the image, a correction coefficient may be calculated for each face area, and the average value may be applied to the RGB values of the entire image data. Further, the correction may be performed by increasing the weight of the face area.

  Next, correction for making the face area an optimal color expression will be described. This corresponds to the processing performed in step 207 described above.

  First, the human face area is obtained from the coordinates of the feature points described in the header of the image data, the area excluding non-skin color parts such as eyes, mouth and hair is determined from the face area, and the YUV format in the area is determined. The signal value of is calculated.

  The gain of the color difference signal is adjusted so as to obtain a preset skin color value. Examples of the adjustment method include a method of changing the color matrix. Since the method for changing the color matrix is known, detailed description thereof is omitted.

  Next, the correction for applying the optimum sharpness to the face area will be described. This corresponds to the processing performed in step 208 described above.

  This is based on the feature point information described in the header, applying moderate sharpness to the contours of organs such as the face, eyes, nose and mouth, and applying a weak low-pass filter to the rest of the face. Image correction is performed to make the face of the selected person more attractive while making wrinkles and blotches less noticeable. The operation flow of this process will be described with reference to FIG.

  The face, eye, nose, and mouth regions of the selected person are acquired from the coordinates of the feature points described in the header of the image data (step 401).

  A band pass filter is applied to the face, nose, and mouth regions to detect the contour (step 402), and the detected contour is enhanced by passing the gain and base clip circuit (step 403).

  In parallel with the process for enhancing the contour, a process for applying a low-pass filter to the entire face region is performed (step 404).

  The output of the gain and base clip circuit and the output of the low pass filter are added using the MIX rate x calculated using the feature point information or the additional information (step 405). At this time, the gain and the output of the base clip circuit and the output of the low-pass filter are added to the regions defined as the contours of the face, eyes, nose, and mouth, and only the output of the low-pass filter is used in the other regions. By this process, the contours of the face, eyes, nose and mouth can be enhanced, and wrinkles and blotches can be made inconspicuous. Also, as the age increases, wrinkles of the corners of the face often stand out, so as shown in FIG. 6, the contour gain is changed according to the age obtained from the additional information described in the header of the image data, Edge enhancement is made softer to make the wrinkles of the corners of the eyes less noticeable.

  The operation for automatically searching for and printing an image in which a specific person is shown has been described above. The above flow can also be applied when two or more specific persons are selected, and even when a plurality of specific persons are designated, if the same process is performed, the printing process can be automatically performed for each person. it can. That is, according to this method, it is possible to automatically perform printing processing for the number of registered persons.

  In the above flow, the feature point information is compared with respect to one image data, the image correction and the printing process are performed, and then the process proceeds to the next image data process. However, the present invention is not limited to this. . First, feature points may be compared with respect to all image data, and all image data determined to contain a specific person may be extracted, and then image correction and print processing may be performed collectively. .

  In the above flow, the feature point information described in the header of the image data is used to determine whether or not the user is the specific person selected by the user. For example, the feature point information is used to laugh. Print only the image data of a person, print only the image data of all the people in the eyes that are open (valid for group photos), or only the image data that is in focus on the face area Can also be printed.

  In addition, from the feature point information described in the header of the image data or additional information accompanying the feature point information, for example, selection by age, selection by gender, selection by facial expression, selection by face orientation, angle of view It is possible to perform image correction and print processing by performing automatic selection under various conditions such as selection based on the ratio of the area occupied by the face.

  The feature point information extraction described above can also be performed on image data already stored in a storage medium, or image data transferred from an external electronic device and stored in a memory or storage medium. .

  The imaging device only obtains image data in RAW format or JPEG format, and the above-described feature point information extraction, identification of a specific person, or image correction is performed by a printer or the like that receives image data from the imaging device. It is also possible to carry out with an apparatus.

  In the first embodiment, image data in JPEG format is generated from image data in RAW format, and feature point information of the image data is detected and described in the header, and image correction and printing are performed using this header information during printing. The configuration for performing the processing has been described.

  In this embodiment, when JPEG image data is generated from RAW image data, a feature point of the image data is detected and image correction is performed, and a JPEG signal that has been subjected to image correction in advance is generated. explain. This is not limited to an imaging apparatus such as a digital camera, but an RAW format image data is transferred from an information processing apparatus such as a PC to which RAW image data obtained by the imaging apparatus has been transferred, or an imaging apparatus. This configuration can also be applied to a printing apparatus such as a printer.

  The contents of the image correction performed in the present embodiment are luminance correction, color reproduction optimization, and sharpness correction in accordance with a person's face, and the detailed correction method is almost the same as the processing described in the first embodiment.

  FIG. 7 shows a parameter setting flow necessary for optimal processing in the development process of RAW format image data. Hereinafter, a flow chart in the case of performing development processing according to the present embodiment will be described.

  First, it is determined whether or not the automatic correction mode is set (step 501). In the automatic correction mode, when a RAW format image data is converted into JPEG format image data, if the face of a specific person is included in the image data, the specific person's face has the optimum brightness and In this mode, image correction is performed so as to achieve color expression. If the automatic correction mode is not set, the RAW format image data is developed by a known method.

  The feature points are detected from the image data in the RAW format, and the detected feature point information is compared with the feature point information stored in advance in the information processing apparatus in the same manner as in the first embodiment. It is determined whether or not the face of a specific person is included in (step 502). If the face of the specific person is not included, the RAW format image data is developed by a known method.

  If it is determined that the face of the specific person is included in the image data, the face area of the specific person is determined from the detected feature point information (step 503).

  Similar to the processing performed in step 206 in FIG. 3, the luminance correction coefficient is obtained using the feature point information so that the luminance information of the face area is optimized (step 504).

  Next, using the feature point information, a color correction matrix is calculated in the same manner as the processing performed in step 207 of FIG. 3 so that the skin color of the face region is optimized (step 505). At this time, a finer correction value can be obtained by considering the gender and age information described in the additional information stored in advance in the information processing apparatus.

  Further, using the feature point information, an edge enhancement parameter for sharpness correction is calculated in the same manner as the processing performed in step 208 in FIG. 3 (step 506). As in the first embodiment, it is preferable to consider the gender and age information described in the additional information. For example, as the age increases, wrinkles in the corners of the eyes and spots on the skin tend to be noticeable, so the edge enhancement of the contour is softened and the area other than the contour of the face area is processed to have a softer image quality. Make Miyaku dull.

  Then, development processing is performed using the obtained luminance correction coefficient, color correction matrix, and edge enhancement parameter (step 507).

  Next, a schematic flow when the image correction shown in FIG. 7 is performed in the digital camera will be described with reference to FIG.

  When an imaging command is given, an imaging device such as a CCD accumulates charges and transfers them (step 601). This output is subjected to blacking processing (step 602), and a white balance gain value is calculated based on the output (step 603).

  If the automatic correction mode is set in the digital camera, the obtained white balance gain value is multiplied only with the face area obtained from the feature point information of the image data to convert it into an RGB signal, and further the face area Color conversion is performed using separate matrices for the other areas and other areas to generate color difference signals (U signal, V signal) (step 604).

  The band of this UV signal is limited by the low-pass filter (step 605), and the color saturation portion is erased (step 606).

  Also, a luminance signal (Y signal) is generated from the output signal of the matrix circuit (step 607), and different edge enhancement processing is performed on the face area and the area other than the face (step 608).

  The image correction described with reference to FIG. 7 is performed during the achromatic processing and edge enhancement processing of the color saturation portion.

  Thereafter, RGB conversion is performed using the generated UV signal and Y signal, and γ correction is performed on this signal. The γ-corrected signal is converted again into a YUV format signal to obtain JPEG format image data. Note that the blackening process, the calculation of the white balance gain value, the conversion to the RGB signal, the achromatization, and the like described above are known image processing methods, and thus detailed description thereof is omitted.

  As described above, in this embodiment, in the development processing for converting RAW format image data into JPEG format image data, a pre-registered person is copied using the header information attached to the RAW format image data. The RAW format image data is automatically searched, and further developed while performing automatic correction so that the face of the person has the optimum luminance level, color reproduction, and resolution. If there are a plurality of persons registered in advance in the same image data, the above process may be repeated so as to be optimal for each person, and the developing process may be performed for the number of registered persons.

  The present embodiment is characterized in that image data is automatically organized or classified on an imaging device such as a digital camera or an information processing device such as a PC using header information of image data in JPEG format or RAW format. .

  For example, feature point information or additional information of image data that has not yet been classified is read out, and based on this, it is assigned to a plurality of arbitrary folders created by the user, or image data is duplicated.

  An example is shown in FIG. There are three folders on the computer. The folder 1 classifies image data that includes Mr. A, the folder 2 classifies image data that includes Mr. B, and the folder 3 includes Mr. C. The image data in which is shown is configured to be classified. Here, when six new image data (image data 1 to 6) are transferred from the digital camera to the PC, the PC compares the header information of these six image data with each folder information, and automatically acquires the image data. Sort out.

  In the example of FIG. 9, Mr. A and Mr. B are shown in the image data 1, only Mr. A is shown in the image data 2, and only the landscape is shown in the image data 3. Mr. B is shown, Mr. C is shown in the image data 5, and Mr. A and Mr. B are shown in the image data 6.

  The feature point information is described in the headers of the image data 1 to 6, and by comparing with each folder information, other image data except the image data 5 in which no one is shown is classified into any folder. . Here, since image data 1 and 6 include Mr. A and Mr. B, the image data is duplicated and classified into both folders.

  As described above, by using the feature point information described in the header attached to the image data, the images can be classified in a short time.

  In this embodiment, the image data is classified using only the feature point information. However, the image data may be classified using additional information attached to the feature point information. For example, FIG. 9B classifies image data using gender information described in the additional information, and FIG. 9C classifies image data using age information described in the additional information. Is.

  As described above, according to the present embodiment, it is possible to classify the image data according to the photographed person without displaying the image data as an image on a display device or the like.

  Although the invention made by the present inventor has been described based on the embodiments, the present invention is not limited to the above-described embodiments, and various modifications can be made without departing from the scope of the invention.

1 is a block diagram illustrating a configuration of an imaging device. The flowchart figure until image data is memorize | stored after the release switch in an imaging device is pushed. The flowchart figure until it searches automatically for the image in which the specific person in the direct print system was reflected, and it prints. The figure which shows the relationship between the area of a face area | region, and a brightness correction coefficient. The flowchart for performing the sharpness correction optimal for a face area. The figure which shows the relationship between age and a sharpness gain. 6 is a flowchart for setting parameters necessary for optimum processing in a development process of RAW format image data. 7 is a flowchart showing a development process of RAW format image data. The figure which shows the example which classifies image data.

Explanation of symbols

DESCRIPTION OF SYMBOLS 11 Imaging lens 12 Aperture 13 Imaging element 14 A / D conversion part 15 Image signal processing part 16 Compression processing part 17 Face detection processing part 18 Memory 19 Storage medium 20 Drive part 21 Aperture control part 22 Lens control part 23 Release switch 24 System control Part

Claims (30)

In an imaging device that obtains image data by converting the output of an imaging device that converts incident light into an electrical signal into a digital value,
An image pickup apparatus that extracts feature point information for specifying a human face included in the image data, and stores the image data with the feature point information. In an imaging device that obtains image data by converting the output of an imaging device that converts incident light into an electrical signal into a digital value,
An image pickup apparatus that reads image data stored in a storage unit, extracts feature point information for specifying a human face included in the image data, and attaches the feature point information to the image data .   The imaging apparatus according to claim 1, wherein the feature point information is face shape information.   The imaging apparatus according to claim 1, wherein the feature point information is information including a face shape and a coordinate position.   5. The imaging apparatus according to claim 3, wherein the feature point information further includes face color information.   Comparison information set for each person is stored, and the person included in the image data and the person corresponding to the comparison information are the same person by comparing the feature point information and the comparison information. The imaging apparatus according to claim 1, wherein the imaging device is determined.   The imaging apparatus according to claim 6, wherein only the image data determined to include the same person is selected from the plurality of image data using the comparison information.   The imaging apparatus according to claim 7, wherein only the selected image data is transmitted to the outside.   The imaging apparatus according to claim 7, wherein only the selected image data is compressed.   6. The face area where a person's face is located is determined using the feature point information, and the image data is processed so that an exposure value of the face area is appropriate. Imaging device.   Using the feature point information, a region having a predetermined part of the person is further determined from the face region, and different processing is performed on the image data between the face region that is determined to have the predetermined part and the other face region. The imaging apparatus according to claim 9, wherein the imaging apparatus is applied.   The image pickup apparatus according to claim 10, further comprising: determining a region indicating an outline from the face region using the feature point information, and performing image processing for edge enhancement only on the region.   The process performed on the image processing is changed according to the size of the face area of the image data or the age information attached to the image data. Imaging device.   The image pickup apparatus according to claim 10, wherein the image data subjected to the image processing is compressed.   The imaging apparatus according to claim 1, wherein a plurality of pieces of image data are classified using the feature point information. In a printing device that prints image data transferred from outside,
Comparison information provided for each individual is stored in advance, feature point information for identifying a person's face included in image data transferred from the outside is extracted, and the image data is attached with the feature point information. And selecting only image data determined to contain the same person from the plurality of image data using comparison information set for each person, and printing the selected image data. A printing apparatus characterized by the above. In a printing device that prints image data transferred from outside,
Comparison information provided for each individual is stored in advance, feature point information for identifying a person's face included in image data transferred from the outside is extracted, and the image data is attached with the feature point information. , A face area where a person's face is located is determined using the feature point information, and image data is processed and printed so that an exposure value of the face area is appropriate. . An image processing apparatus that compresses image data obtained by the imaging apparatus according to claim 1,
An image processing apparatus that selects and compresses image data in which the attached feature point information satisfies a predetermined condition from a plurality of image data.   Comparison information set for each person is stored, and by comparing the comparison information with the feature point information attached to the image data, a person included in the image data and a person corresponding to the comparison information are obtained. The image processing apparatus according to claim 18, wherein it is determined whether or not they are the same person.   The image processing apparatus according to claim 19, wherein only the image data determined to include the same person is selected from a plurality of image data using the comparison information. An image processing apparatus that compresses image data obtained by the imaging apparatus according to claim 1,
A face area where a person's face is located is determined using feature point information attached to the image data, and the image data is processed and compressed so that an exposure value of the face area is appropriate. Image processing device.   Using the feature point information, a region having a predetermined part of the person is further determined from the face region, and different processing is performed on the image data between the face region that is determined to have the predetermined part and the other face region. The image processing apparatus according to claim 21, wherein the image processing apparatus is applied.   23. The image processing apparatus according to claim 22, further comprising: determining a region indicating an outline from a face region using the feature point information, and performing image processing for edge enhancement only on the region.   The process performed on the image processing is changed according to the size of the face area of the image data or age information attached to the image data. Image processing device.   Reading image data from a storage means, extracting feature point information for specifying a person's face included in the read image data, and storing the image data accompanied with the feature point information. A featured program.   The program according to claim 25, wherein the feature point information includes information related to a face shape.   By reading the comparison information set for each person from the storage means and comparing the feature point information with the comparison information, the person included in the image data and the person corresponding to the comparison information are the same person. 27. The program according to claim 25 or 26, wherein the program is determined.   28. The program according to claim 27, wherein only the image data determined to contain the same person is selected from a plurality of image data using the comparison information.   27. The program according to claim 25 or 26, wherein a face area having a human face is determined using the feature point information, and image data is processed so that an exposure value of the face area is appropriate. .   27. The program according to claim 25 or 26, wherein a plurality of pieces of image data are classified using the feature point information.

Images