JP2006085266A - Digital picture distribution system and digital picture distribution method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a digital picture distribution system and a digital picture distribution method increasing convenience by facilitating users. <P>SOLUTION: The digital picture distribution system 1 includes: a storage device 30 for storing picture data 31; a processor 10 connected to the storage device 30; and a program 20 executed by the processor 10. The processor 10 operates in the following way according to the instructions of the program 20: the processor 10 detects face areas RT covering persons' faces shown within a picture shown by the picture data 31, and extracts characteristic quantities about the face areas RT; based on the characteristic quantities extracted, the processor 10 then selects from among the plurality of face areas RT the face area RT that shows the person to which the picture should be distributed as the selected face area; the processor 10 then executes a process to distribute the picture to the person shown in the selected face area. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to a technique for processing a digital picture, and particularly to a system for distributing a digital picture and a program used in the system.

  Conventionally, in order to reprint and distribute photographs, the following procedure has been required. First, the user looks at the completed photo and identifies the person in the photo. Subsequently, the user determines the number of copies by counting the number of subjects. Then, the user duplicates (prints) the same number of photos and distributes the duplicated photos to each person. This procedure is the same when distributing digital pictures taken with a digital camera or the like.

  There is a demand for a technique that can reduce the user's trouble in copying and distributing photographs and improve convenience.

  As a related technique, a “face recognition technique” for recognizing a human face in a digital image is known. In general, the face recognition processing includes (1) a step of detecting an area (face area) where a face is captured from an image, and (2) a “facial feature (facial feature) indicating a detected face feature. ) ”And (3) a step of referring to the database and collating who the face having the face feature value is. The (1) detection step is performed, for example, by searching for an arrangement pattern of both eyes / nose / mouth. The face feature amount indicates information such as the contour of the face and the shape and relative arrangement of parts such as eyes, nose and mouth. Technologies relating to such face recognition processing are disclosed in, for example, Patent Document 1, Patent Document 2, Patent Document 3, and Patent Document 4.

  Patent Document 5 discloses a technique for organizing images and searching for images using such face recognition technology. The image processing apparatus according to Patent Document 5 includes a registration unit that registers a first face image serving as a search reference as a reference image, an extraction unit that extracts a second face image from an input image, and a second face image. Search means for searching for a reference image showing the same person, storage means for storing at least one of the name and shooting year of the person added to the reference image corresponding to the search together with the second face image, and the corresponding reference Display means for displaying the image in association with the second face image.

JP 7-311833 A JP 2001-283224 A JP 2003-187229 A JP 2003-296735 A JP 2003-150932 A

  An object of the present invention is to provide a digital picture distribution system and a digital picture distribution method that can reduce the user's effort and improve convenience.

  [Means for Solving the Problems] will be described below using the numbers and symbols used in [Best Mode for Carrying Out the Invention]. These numbers and symbols are added in parentheses in order to clarify the correspondence between the description of [Claims] and [Best Mode for Carrying Out the Invention]. However, these numbers and symbols should not be used for the interpretation of the technical scope of the invention described in [Claims].

  A digital picture distribution system (1) according to the present invention includes a storage device (30) for storing picture data (31), an arithmetic processing device (10) connected to the storage device (30), and an arithmetic operation thereof. And a program (20) executed by the processing device (10). The arithmetic processing unit (10) operates as follows according to the instructions of the program (20). That is, the arithmetic processing unit (10) detects a face area (RT) in which a person's face is shown in the picture indicated by the picture data (31), and extracts a “feature” related to the face area (RT). . Next, the arithmetic processing unit (10) selects a face area (RT) in which a person to whom a photo is to be distributed is selected from the plurality of face areas (RT) based on the extracted feature amount. Select as region. Next, the arithmetic processing unit (10) executes processing for distributing a photograph to a person shown in the selected face area.

  The feature amount includes “size feature amount (s)” indicating the size of the face region (RT). At this time, the arithmetic processing unit (10) selects the face area (RT) as the selected face area when the size is equal to or larger than a predetermined threshold value, and if the size is smaller than the predetermined threshold value, the face area (RT) is not selected. Further, the “size feature (s)” may indicate a ratio of the size of the face region (RT) to the frame (FL) of the photograph. At this time, the arithmetic processing unit (10) selects the face area (RT) as the selected face area when the ratio is equal to or greater than a predetermined threshold value, and selects the face area (RT) when the ratio is smaller than the predetermined threshold value. ) Is not selected.

  The feature amount includes “coordinate feature amount (x, y)” indicating the position of the face region (RT) in the photograph. At this time, when the position is within a predetermined area (CR) including the center of the photograph, the arithmetic processing unit (10) selects the face area (RT) as the selected face area, and the position is determined to be the predetermined area. If it is outside the region (CR), the face region (RT) is not selected.

  In this digital picture distribution system (1), the arithmetic processing unit (10) specifies a person appearing in the selected face area, and analyzes result data (corresponding to “identified person” and “photograph” in association with each other). 33) is stored in the storage device (30). Then, the arithmetic processing device (10) executes processing for distributing a photograph to the specified person based on the analysis result data (33). Specifically, the digital picture distribution system (1) further includes a database (40) connected to the arithmetic processing unit (10). The database (40) includes a person's name (NAME). The contact information (ADD, EMAIL) and the “face feature (FACE)” indicating the feature of the face are stored in association with each other. The extracted feature amount also includes a face feature amount (f). At this time, the arithmetic processing unit (10) refers to the database (40) and compares the extracted face feature quantity (f) with the face feature quantity (FACE) indicated by the database (40), thereby selecting the selected face. Identify the person in the area.

  In this digital picture distribution system (1), the contact information stored in the database (40) includes an e-mail address (EMAIL) and an address (ADD). The arithmetic processing unit (10) acquires the e-mail address (EMAIL) of the specified person by referring to the database (40), and sends picture data (31) to the e-mail address (EMAIL). You may send it. The arithmetic processing unit (10) acquires the number of persons associated with the photograph by referring to the analysis result data (33), and displays information indicating the number of persons together with the picture data (31) on the network ( 200) to the printer (210). The arithmetic processing unit (10) may store the picture data (31) in the portable medium (150). The arithmetic processing unit (10) acquires the number of persons associated with the photograph by referring to the analysis result data (33), and controls the printer (140) to print the number of photographs corresponding to the number of persons. May be. The processor (10) may refer to the database (40) and control the printer (140) to print a postal label indicating the name (NAME) and address (ADD) of the identified person.

  The program (20) according to the present invention includes: (a) a step of detecting a face region (RT) in which a person's face is shown in a photograph indicated by picture data (31); and (b) the face region (RT). A step of extracting a “feature amount” relating to: (c) based on the feature amount, a face area (RT) in which a person to whom a photograph is to be distributed is selected from a plurality of face areas (RT) The computer is caused to execute a step of selecting as an area and (d) a step of executing a process for distributing a photograph to a person in the selected face area.

  The digital picture distribution method according to the present invention is a digital picture distribution method by the digital picture distribution system (1) described above. In this distribution method, (A) the arithmetic processing unit (10) reads the picture data (31) indicating a photograph, and (B) the arithmetic processing unit (10) adds “face detection” to the picture data (31). A step of detecting a face region (RT) in which a person's face is shown in the photograph by performing “processing”, and (C) the arithmetic processing unit (10) performs “feature amount” relating to the face region (RT). And (D) the arithmetic processing unit (10), based on the feature amount, out of a plurality of face regions (RT), a face region (RT) in which a person to whom a photograph is to be distributed is shown And (E) the arithmetic processing unit (10) executes a process for distributing a photograph to a person in the selected face area.

  In the step (C), the extracted feature quantity may include a “size feature quantity (s)” indicating the size of the face region (RT). At this time, in the step (D), when the size is equal to or larger than a predetermined threshold, the arithmetic processing unit (10) selects the face region (RT) as the selected face region, and the size is the predetermined threshold. If it is smaller, the face area (RT) is not selected. The extracted feature amount may indicate the ratio of the size of the face region (RT) to the frame (FL) of the photo. At this time, in the step (D), when the ratio is equal to or larger than a predetermined threshold, the arithmetic processing unit (10) selects the face area (RT) as the selected face area, and the ratio is smaller than the predetermined threshold. In this case, the face area (RT) is not selected. The extracted feature quantity may include “coordinate feature quantity (x, y)” indicating the position of the face region (RT) in the photograph. At this time, in the step (D), when the position is within a predetermined area (CR) including the center of the photograph, the arithmetic processing unit (10) selects the face area (RT) as the selected face area. If the position is outside the predetermined area (CR), the face area (RT) is not selected.

  In the step (C), the feature quantity extracted includes a face feature quantity (f). At this time, in the step (E), (E1) the arithmetic processing unit (10) refers to the database (40), and the extracted facial feature quantity (f) and the facial feature quantity (FACE) indicated by the database (40). ) To identify a person in the selected face area, and (E2) a step in which the arithmetic processing unit (10) executes a process for distributing a photograph to the identified person And.

  According to the digital picture distribution system and the digital picture distribution method according to the present invention, the user's labor is reduced and the convenience is improved.

  A digital picture distribution system according to the present invention, a program used for the digital picture distribution method, and a digital picture distribution method will be described with reference to the accompanying drawings.

(Constitution)
FIG. 1 is a block diagram showing a configuration of a digital picture distribution system according to an embodiment of the present invention. The digital picture distribution system 1 includes an arithmetic processing device 10, picture processing software 20, a storage device 30, a face / person information database 40, an input device 50, and a display device 60.

  The picture processing software 20 is a computer program executed by the arithmetic processing device 10, and the arithmetic processing device 10 operates according to instructions of the picture processing software 20. The picture processing software 20 includes a reading unit 21, a face detection unit 22, a person analysis unit 23, and a distribution processing unit 24. As will be described in detail later, the reading unit 21 realizes a function of reading picture data 31 that is data of a digital picture (photograph). The face detection unit 22 performs a “face detection process” on the picture data 31 to detect a region (face region) in which a face is reflected from the photograph, and further extracts a “feature” related to the face region. Realize the function to do. Thereby, feature amount data 32 is created that indicates the detected face area and the extracted feature amount in association with each other. The person analysis unit 23 realizes a function of selecting or specifying a person to whom a photograph is to be distributed by analyzing the feature amount data 32. Thereby, analysis result data 33 indicating the result of the analysis is created. The distribution processing unit 24 realizes a function for performing various processes for distributing photographs based on the analysis result data 33.

  The storage device 30 is connected to the arithmetic processing device 10. Examples of the storage device 30 include memories such as DRAM and SRAM. The storage device 30 can store the above-described picture data 31, feature data 32, analysis result data 33, and the like.

  The face / person information database 40 is connected to the arithmetic processing apparatus 10. The face / person information database 40 is stored in, for example, a hard disk. FIG. 2 is a diagram conceptually showing the contents of the face / person information database 40 according to the present embodiment. As shown in FIG. 2, the face / person information database 40 stores a person's name, face feature amount, and contact information (address or e-mail address) in association with each other. Here, the “face feature amount” indicates information such as the contour of the face of the person, the shape of the parts such as eyes, nose, and mouth, and relative arrangement. For example, a name indicated by “No. 1” is associated with a name NAME1, a face feature value FACE1, an address ADD1, and a mail address EMAIL1. The face / person information database 40 may have an acquaintance / other person flag. For example, if a person is an acquaintance of the user, the acquaintance / others flag is set. For example, in FIG. 2, the person indicated by “No. 4” is another person, and information other than the face feature amount is not associated. As will be described later, the face / person information database 40 can be updated.

  The input device 50 and the display device 60 are connected to the arithmetic processing device 10. Examples of the input device 50 include a keyboard and a mouse. The display device 60 presents information to the user.

  The digital picture distribution system 1 can be connected to the network 200 via the network interface 70, and can be connected to the net print dealer 210 and other terminals (PCs) 220 on the network 200. The net print trader 210 is a trader that accepts orders for printing digital pictures online. Further, by using the input / output interface 80, a digital camera 110, a portable terminal 120, a scanner 130, a printer 140, a portable medium (portable media drive) 150, and the like can be connected to the digital picture distribution system 1. Is possible. An example of the mobile terminal 120 is a mobile phone with a digital camera. Examples of the portable medium (portable media drive) 150 include a memory card, a drive for writing on a CD-R, and a DVD.

(Operation)
FIG. 3 is a diagram showing an overall flow of the operation of the digital picture distribution system 1. First, the reading unit 21 reads the picture data 31 (step S10). Next, the face detection unit 22 performs “face detection processing” on the picture data 31 (step S20). Next, the person analysis unit 23 performs person analysis processing on the feature amount data 32 (step S30). Next, the distribution processing unit 24 performs a process for distributing the photograph (step S40). Hereinafter, each step will be described in detail.

Step S10 (reading process):
First, a plurality of picture data 31 indicating a digital picture (photograph) is read. The picture data 31 is read from the digital camera 110, the portable terminal 120, the memory card (portable medium 150), or the like via the input / output interface 80. Further, the photograph 130 may be read by scanning a photograph printed by the scanner 130. In addition, the picture data 31 may be read from another PC 220 via the network 200. The plurality of read picture data 31 is stored in the storage device 30.

Step S20 (face detection process):
Next, a well-known “face detection process” is performed on the picture data 31. That is, a face region (hereinafter referred to as “target RT”) in which a person's face is reflected is detected from the picture indicated by the picture data 31. This is performed, for example, by searching for an arrangement pattern of both eyes / nose / mouth in the image. Here, a plurality of target RTs may be detected for one picture data 31, and there may be cases where no target RT is detected at all. Subsequently, a “feature amount” (hereinafter referred to as “target feature amount”) regarding each detected target is extracted. As a result, feature amount data 32 is created that indicates the detected target and the extracted target feature amount in association with each other. The feature amount data 32 is stored in the storage device 30.

FIG. 4 is a conceptual diagram showing the contents of the feature amount data 32. As shown in FIG. 4, it is assumed that four targets RT1 to RT4 are detected for a certain picture data 31. Each target RT is associated with a target feature amount. This target feature amount includes, for example, a plurality of feature amounts as follows:
(1) “Coordinate feature amount x, y” indicating the coordinates (center position) of the target RT in the frame FL;
(2) “Size feature s” indicating the size of the target RT or the ratio of the size of the target RT to the frame FL;
(3) “Inclination feature value θ” indicating the inclination of the face in the target RT;
(4) “Face feature amount f” indicating the feature of the face in the target RT.
The face feature amount f indicates information such as the contour of the face, the shape of parts such as eyes, nose, and mouth, and the relative arrangement thereof. In FIG. 4, for example, the target RT2 is associated with the target feature amount (x2, y2, s2, θ2, f2). The same applies to other target RTs.

  Such processing can be performed on a plurality of photographs (picture data 31). Among the plurality of photographs, there may be a photograph in which a person is not shown like a landscape photograph. Or there may be a photograph in which an unspecified number of people are reflected, such as a photograph of an urban area. In other words, there may be a photograph that does not require reprinting. In this case, by using the input device 50, the user can designate a photo that needs to be reprinted from a plurality of photos displayed on the display device 60. Then, the above-described processing is performed only on the designated photograph. Alternatively, when the target RT is not detected at all as a result of the face detection process, the face detection unit 22 may automatically determine that the photograph is a landscape photograph and is not required to be reprinted. Further, when the number of targets RT equal to or greater than a predetermined threshold is detected, the face detection unit 22 may automatically determine that the photograph is not required to be reprinted. If it is determined that one photo is not needed, the process automatically moves to the next photo. In this way, it is possible to automatically and sequentially execute processing for a plurality of photographs.

Step S30 (person analysis processing):
Next, person analysis processing is performed based on the above-described feature amount data 32. The purpose of this person analysis process is as follows. That is, there are often cases where strangers are reflected in the background of a photograph in which a user or an acquaintance of the user is reflected. However, in the above-described step S20, there is a possibility that such other person's face is also detected as the target RT. However, it is not necessary to distribute the reprint of the photo to the stranger. Therefore, the first purpose of the person analysis processing is to automatically select only the necessary target RT from the detected plurality of target RTs. The second purpose of the person analysis process is to identify a person shown in the selected target RT. In other words, this person analysis process is necessary because the photograph is reprinted and distributed, and is realized as follows in the present invention.

  FIG. 5 is a conceptual diagram showing the data flow in this person analysis process. FIG. 6 is a flowchart showing the person analysis process in detail. First, the person analysis unit 23 reads the created feature data 32. Then, the person analysis unit 23 selects a target RT (hereinafter referred to as “selected target”) in which a person to whom a photograph is to be distributed is shown from at least one target RT detected in step S20. This selection process is performed based on the target feature amount indicated in the feature amount data 32. As a result, the selection target to be subjected to the subsequent person identification process is determined (step S31).

  Specifically, this step S31 is executed as follows. In a photo, you and your acquaintance are more likely to appear larger than others in the background. As described above, the “size feature amount s” indicated in the feature amount data 32 indicates the size of the target RT or the ratio of the size of the target RT to the picture frame FL. Therefore, the person analysis unit 23 can select the selection target based on the “size feature quantity s”. In addition, since the face shown in the target RT may face obliquely, the size feature quantity s corrected using the tilt feature quantity θ may be used.

  For example, in the example shown in FIG. 4, the sizes of the targets RT1 and RT4 are smaller than the sizes of the targets RT2 and RT3. The person analysis unit 23 refers to the size feature amounts s1 to s4 and determines whether the size feature amounts s1 to s4 are larger or smaller than a predetermined threshold. This predetermined threshold can be freely set by the user using the input device 50. For example, it is assumed that the feature amounts s1 and s4 are smaller than the predetermined threshold value and the feature amounts s2 and s3 are determined to be larger than the predetermined threshold value. In this case, the person shown in the targets RT1 and RT4 is likely to be another person, and the targets RT1 and RT4 may be excluded. Therefore, the person analysis unit 23 selects the targets RT2 and RT3 as selection targets.

  Also, in a certain photo, it is highly likely that the person or acquaintance is in the center of the photo, and that other people are likely to be in the periphery of the photo. As described above, “coordinate feature amounts x, y” indicated in the feature amount data 32 indicate the coordinates (center position) of the target RT in the frame FL. Therefore, the person analysis unit 23 can select the selection target based on the “coordinate feature amounts x, y”.

  For example, in the example shown in FIG. 4, the targets RT2 and RT3 are located at the center of the targets RT1 and RT4. The person analysis unit 23 refers to the coordinate feature amounts (x1, y1) to (x4, y4), and determines whether the targets RT1 to RT4 are inside or outside the predetermined central region CR. The predetermined central region CR is a region including the center of the picture frame FL and having a predetermined size. The predetermined size can be freely set by the user using the input device 50. For example, it is assumed that it is determined that the targets RT1 and RT4 are located outside the central region CR and the targets RT2 and RT3 are located within the central region CR. In this case, the person shown in the targets RT1 and RT4 is likely to be another person, and the targets RT1 and RT4 may be excluded. Therefore, the person analysis unit 23 selects the targets RT2 and RT3 as selection targets.

  Alternatively, in a certain photo, it is highly likely that the face of the other person in the photo is facing diagonally rather than in front. Therefore, the person analysis unit 23 can select the selection target based on the “tilt feature amount θ”. Or, in a certain photo, the face of another person is likely to be out of focus. Therefore, the target feature value described above includes a focus feature value indicating defocusing, and the person analysis unit 23 can select a selection target based on the focus feature value. The person analysis unit 23 may perform determination using a plurality of feature amounts among the target feature amounts.

  Thus, this step S31 is a step that is necessary because the distribution process for distributing the photograph (step S40; see FIG. 3) is performed later.

  Next, “person identification processing” for identifying a person shown in the selected target is executed (step S32). At this time, as shown in FIG. 5, the person analysis unit 23 refers to the face / person information database 40. As shown in FIG. 2, in the face / person information database 40, a person is associated with a “face feature amount FACE”. Therefore, the person analysis unit 23 can compare “the face feature amount f indicated in the feature amount data 32” with “the face feature amount FACE indicated in the face / person information data 40”. Thereby, face collation is performed, and it becomes possible to specify a person shown in the selected target. For example, when the face feature amount f2 (see FIG. 4) matches the face feature amount FACE1 (see FIG. 2), the person shown in the target RT2 is specified as the person indicated by “No. 1”. When the face feature amount f3 matches the face feature amount FACE3, the person shown in the target RT3 is specified as the person indicated by “No. 3”. In this way, a certain photo is associated with the person to whom the photo is to be distributed.

  If there is no unidentifiable target among the at least one selected target (step S33; No), “analysis result data 33” indicating the identified person and the photograph in association with each other is created (step S36). ). When there is an unidentifiable target among the selected targets (step S33; Yes), the person analysis unit 23 presents that fact on the display device 60 and prompts the user to input information regarding the target. When the notification is made, the user can input the name, address, and mail address of the person shown on the target using the input device 50. In addition, when the selection target is erroneously determined, the user can perform correction using the input device 50.

  Further, the user can newly register information on the target in the face / person information database 40. For example, the person analysis unit 23 displays buttons such as “register” and “not register” on the display device 60. When the user designates “registration” (step S34; Yes), the information (name, address, e-mail address, etc.) input as described above together with the face feature quantity (f, FACE) and the face / person information database 40 Newly added. Thereby, the face / person information database 40 is updated (step S35). At this time, an acquaintance / other person flag can be designated, and only the face feature amount of the other person can be registered in the face / person information database 40 (see FIG. 2). When the user designates “not register” (step S34; No), the update is not performed.

  The above processing is executed for each of the plurality of picture data 31. That is, each picture data 31 is associated with a person shown in the selected target in the photo. Then, “analysis result data 33” indicating such correspondence is created (step S36).

  FIG. 7 is a conceptual diagram showing the contents of the analysis result data 33. As shown in FIG. 7, each picture data 31 is associated with the number of selected targets (number of people) and information of the specified person (person). The number of selected targets (number of people) represents the number of extra prints. As the identified person information (person), for example, a name NAME registered in the face / person information database 40 is shown. For example, “3” is associated with “Picture” and “NAME1, NAME2, NAME3” are associated with “Picture” in a certain picture PIC1. Similarly, “persons” and “persons” are associated with the other pictures PIC2 and PIC3. The created analysis result data 33 is stored in the storage device 30.

Step S40 (distribution processing):
Next, based on the created analysis result data 33, a process for distributing a photograph to each person indicated in the analysis result data 33 is executed. FIG. 8 is a conceptual diagram showing the flow of data in this distribution process. FIG. 9 is a flowchart showing details of the distribution process.

  First, as illustrated in FIG. 8, the distribution processing unit 24 reads picture data 31 and analysis result data 33. The contents of the analysis result data 33 are displayed on the display device 60 together with the photograph indicated by the picture data 31. The user refers to the display and uses the input device 50 to change the analysis result data 33 and specify the distribution method (step S401).

  The change of the analysis result data 33 is for improving the degree of freedom of the user, and may be executed only when necessary. For example, it may not be necessary to send a photo to all of the persons identified in step S30. Or you may send a photo to someone who is not in the picture. Therefore, the user can use the input device 50 to add or delete a person to be distributed as appropriate. Thereby, the analysis result data 33 which is the basis of the distribution process is changed.

  In specifying the distribution method, the display device 60 displays buttons such as “E-mail”, “Net print dealer”, “Portable media”, and “Print”, for example. The user can select a desired distribution method from among them by using the input device 50. As an option, a “mail” button for designating whether or not to mail the photograph may be provided. If the user wants to mail the photo, he checks the “mail” button.

  When “E-mail” is designated, the distribution processing unit 24 compresses the picture data 31 to a file size suitable for mail transmission (step S411). This compression process may be performed as necessary. Next, the distribution processing unit 24 refers to the face / person information database 40 and acquires the mail address EMAIL of each person (step S412). For example, consider the case of the analysis result data 33 shown in FIG. The distribution processing unit 24 acquires, for example, the mail address MAILIL1 of the person indicated by “NAME1”. According to the analysis result data 33, the person is shown in three pictures PIC1 to PIC3. Therefore, the distribution processing unit 24 transmits an email attached with the three picture data 31 corresponding to the three pictures PIC1 to PIC3 to the email address EMAIL1 (step S413). These three picture data 31 may be compressed in step S411. The same applies to other persons (NAME2, NAME3). In this way, necessary picture data 31 is distributed to each person.

  When “net print dealer” is designated, the distribution processing unit 24 refers to the “number of people” information in the analysis result data 33. This number information means the number of extra prints as it is. Therefore, the distribution processing unit 24 adds this number information (printed number information) to each picture data 31 (step S421). For example, “3 sheets” is added to the picture data 31 corresponding to the picture PIC1 as the number-of-prints information. Then, the distribution processing unit 24 transmits the picture data 31 and the extra print number information to the net print dealer 210 via the network 200. Thereby, the print order is completed (step S422).

  When “portable medium” is designated, the picture data 31 is written to the portable medium 150 (step S431). For example, when a CD-R drive is connected to the digital picture distribution system 1, picture data 31 is written to the CD-R. Here, it is sufficient to prepare as many CD-Rs as the number of persons indicated in the analysis result data 33. For example, in the case of the analysis result data 33 shown in FIG. 7, three CD-Rs are prepared for distributing photographs to each of three persons (NAME1 to NAME3). Then, for the person indicated by NAME1, the distribution processing unit 24 writes the three picture data 31 corresponding to the three pictures PIC1 to PIC3 to one CD-R. The same applies to other persons (NAME2, NAME3). In this way, picture data 31 necessary for each person is written on the CD-R. The portable medium 150 includes a memory card and the like.

  When “print” is designated, the distribution processing unit 24 refers to the “number of people” information in the analysis result data 33. This number information means the number of extra prints as it is. Therefore, the distribution processing unit 24 controls the printer 140 so as to print the photo by the number of extra prints (step S441). For example, in the example shown in FIG. 7, three pictures, one picture, and two pictures PIC1 to PIC3 are printed, respectively. Here, the photos may be sorted and printed for each person.

  If “mail” is designated as an option (step S450; Yes), the distribution processing unit 24 refers to the face / person information database 40 and acquires the name NAME and address ADD of each person (see FIG. 2). Then, the distribution processing unit 24 controls the printer so as to print a postal label indicating the name and address (step S451). As a result, a postal label for mailing the printed photograph or the portable medium 150 storing the data to each person is automatically obtained.

  As described above, according to the digital picture distribution technique of the present invention, the user's effort is reduced and the convenience is improved. The user needs to specify at least the distribution method. The digital picture distribution system 1 automatically analyzes the picture data 31 and automatically executes processing for distributing photographs. In particular, in the person analysis process (step S30) described above, the selected target can be automatically determined from the targets RT detected by the face detection process. This is important for automatically reprinting and distributing photos.

FIG. 1 is a block diagram showing a configuration of a digital picture distribution system according to an embodiment of the present invention. FIG. 2 is a conceptual diagram showing the contents of the face / person information database according to the embodiment of the present invention. FIG. 3 is a flowchart showing a digital picture distribution method according to the embodiment of the present invention. FIG. 4 is a conceptual diagram showing the contents of the feature amount data according to the embodiment of the present invention. FIG. 5 is a block diagram for explaining person analysis processing by the digital picture distribution system according to the embodiment of the present invention. FIG. 6 is a flowchart showing person analysis processing according to the embodiment of the present invention. FIG. 7 is a conceptual diagram showing the contents of the analysis result data according to the embodiment of the present invention. FIG. 8 is a block diagram for explaining distribution processing by the digital picture distribution system according to the embodiment of the present invention. FIG. 9 is a flowchart showing distribution processing according to the embodiment of the present invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Digital picture distribution system 10 Arithmetic processing apparatus 20 Picture processing software 21 Reading part 22 Face detection part 23 Person analysis part 24 Distribution processing part 30 Storage device 31 Picture data 32 Feature-value data 33 Analysis result data 40 Face and person information database 50 Input Device 60 Display device 70 Network interface 80 Input / output interface 110 Digital camera 120 Mobile terminal 130 Scanner 140 Printer 150 Portable media 200 Network 210 Net print trader 220 PC

Claims (20)

A storage device for storing picture data;
An arithmetic processing unit connected to the storage device;
A program executed by the arithmetic processing unit,
The arithmetic processing unit operates according to instructions of the program,
Detecting a face area in which a person's face is reflected in the picture indicated by the picture data;
Extracting feature quantities related to the face area;
Based on the feature amount, a face area in which a person to whom the photo is to be distributed is selected as a selected face area among the plurality of face areas,
A digital picture distribution system that executes processing for distributing the photograph to a person in the selected face area. The digital picture distribution system according to claim 1,
The feature amount indicates the size of the face area,
The arithmetic processing unit selects the face area as the selected face area when the size is equal to or larger than a predetermined threshold, and does not select the face area when the size is smaller than the predetermined threshold. Digital picture distribution system. The digital picture distribution system according to claim 1,
The feature amount indicates a ratio of the size of the face area to the frame of the photo,
The digital processing distribution system, wherein the arithmetic processing unit selects the face area as the selected face area when the ratio is equal to or greater than a predetermined threshold, and does not select the face area when the ratio is smaller than the predetermined threshold. The digital picture distribution system according to claim 1,
The feature amount indicates a position of the face region in the photograph,
The arithmetic processing unit selects the face area as the selected face area when the position is within a predetermined area including the center of the photograph, and when the position is outside the predetermined area, Digital picture distribution system that does not select a face area. A digital picture distribution system according to any one of claims 1 to 4,
The arithmetic processing unit includes:
The person in the selected face area is identified, analysis result data indicating the identified person and the photograph in association with each other is stored in the storage device, and the specified person is identified based on the analysis result data. A digital picture distribution system for executing processing for distributing the photograph to a person. The digital picture distribution system according to claim 5,
Furthermore, it comprises a database connected to the arithmetic processing unit,
The database stores a person's name, a contact information, and a facial feature amount indicating a facial feature in association with each other,
The extracted feature amount includes the face feature amount,
The arithmetic processing device refers to the database, and compares the extracted facial feature quantity with the facial feature quantity indicated by the database to identify a person in the selected face area. Digital picture distribution system . The digital picture distribution system according to claim 6,
The contact includes an email address;
The digital processing distribution system, wherein the arithmetic processing unit acquires the e-mail address of the identified person by referring to the database, and transmits the picture data to the e-mail address. The digital picture distribution system according to claim 6,
The arithmetic processing unit obtains the number of persons associated with the photograph by referring to the analysis result data, and sends information indicating the number of persons together with the picture data to a printer through a network. Send digital picture distribution system. The digital picture distribution system according to claim 6,
The arithmetic processing unit is a digital picture distribution system that stores the picture data in a portable medium. The digital picture distribution system according to claim 6,
The arithmetic processing apparatus obtains the number of persons associated with the photograph by referring to the analysis result data, and controls a printer to print the photographs by the number of persons. Digital picture distribution system . The digital picture distribution system according to claim 9 or 10,
The contact information includes an address,
The digital processing distribution system, wherein the arithmetic processing unit refers to the database and controls a printer to print a postal label indicating the name and the address of the identified person. (A) detecting a face region in which a person's face is shown in a picture indicated by picture data;
(B) extracting a feature amount related to the face region;
(C) selecting, as a selected face area, a face area in which a person to whom the photograph is to be distributed is shown among the plurality of face areas based on the feature amount;
(D) A program for causing a computer to execute a process of distributing the photograph to a person in the selected face area. A program according to claim 12,
The extracted feature amount indicates the size of the face region,
In the step (c),
A program for selecting the face area as the selected face area when the size is equal to or larger than a predetermined threshold value, and causing the computer to execute a step of not selecting the face area when the size is smaller than the predetermined threshold value. A program according to claim 12,
The extracted feature amount indicates a ratio of the size of the face area to the frame of the photo,
In the step (c),
A program for selecting the face area as the selected face area when the ratio is equal to or greater than a predetermined threshold, and causing the computer to execute a step of not selecting the face area when the ratio is smaller than the predetermined threshold. A program according to claim 12,
The extracted feature amount indicates the position of the face region in the photograph,
In the step (c),
When the position is within a predetermined area including the center of the digital picture, the face area is selected as the selected face area, and when the position is outside the predetermined area, the face area is not selected. A program that causes a computer to execute steps. A digital picture distribution method in a digital picture distribution system comprising an arithmetic processing unit that operates according to instructions of a computer program,
(A) the arithmetic processing unit reads picture data indicating a photograph;
(B) The arithmetic processing unit detects a face region in which a human face is reflected in the photograph by performing a face detection process on the picture data;
(C) the arithmetic processing unit extracting a feature amount related to the face area;
(D) The arithmetic processing unit selects, as a selected face area, a face area in which a person to whom the photograph is to be distributed is shown among the plurality of face areas based on the feature amount;
(E) The digital processing distribution method comprising: a step in which the arithmetic processing unit executes a process for distributing the photograph to a person in the selected face area. The digital picture distribution method according to claim 16, comprising:
In the step (C), the extracted feature amount indicates the size of the face area,
In the step (D), the arithmetic processing unit selects the face area as the selected face area when the size is equal to or larger than a predetermined threshold, and when the size is smaller than the predetermined threshold, Digital picture distribution method that does not select an area. The digital picture distribution method according to claim 16, comprising:
In the step (C), the extracted feature amount indicates a ratio of the size of the face area to the frame of the photograph,
In the step (D), the arithmetic processing unit selects the face area as the selected face area when the ratio is equal to or larger than a predetermined threshold, and selects the face area when the ratio is smaller than the predetermined threshold. Do not select Digital picture distribution method. The digital picture distribution method according to claim 16, comprising:
In the step (C), the extracted feature amount indicates the position of the face area in the photograph,
In the step (D), when the position is within a predetermined area including the center of the photograph, the arithmetic processing unit selects the face area as the selected face area, and the position is the predetermined area. A digital picture distribution method in which the face area is not selected if it is outside. A digital picture distribution method according to any one of claims 16 to 19,
The digital picture distribution system further includes a database that stores a person's name, a contact information, and a facial feature amount indicating a facial feature in association with each other,
In the step (C), the extracted feature quantity includes the face feature quantity,
The step (E)
(E1) The arithmetic processing device refers to the database, and compares the extracted face feature quantity with the face feature quantity indicated by the database to identify a person appearing in the selected face area When,
(E2) A digital picture distribution method, comprising: the arithmetic processing unit executing a process for distributing the photograph to the specified person.

Images