JP2006113741A - Face detecting device, face detecting method and face detecting program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a face detecting device capable of identifying a plurality of face images in a photographic image, and to provide a face detecting method and a face detecting program. <P>SOLUTION: A management number acquisition part 12 acquires a management number from information (including a start code, management number and end code or the like) read from an IC card 6 by an IC card reader 3. The management number is unique identification information assigned to every IC card 6. A link part 13 stores face management information obtained by associating the specified face position of a face detecting part 11 with the management number read from the IC card 6 in a management DB 4. Image processing for making it difficult to browse any face section other than the face section corresponding to the designated management number may be performed. Alternately, a face image corresponding to the designated management number may be arranged in the specific composite region of a template image. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to a face detection apparatus, a face detection method, and a face detection program, and more particularly, to an apparatus, method, and program for detecting a human face portion.

2. Description of the Related Art Conventionally, a technique for identifying which person a face image that is an image of a face portion of a person image has been developed. For example, according to Patent Document 1, the foreign object detection unit detects a foreign object (for example, eye patch) included in the face image of the person to be recognized photographed by the camera, using the template image stored in the template image DB. The face image correction unit deletes a part corresponding to the foreign object detected by the foreign object detection unit from the face image, and performs a process of correcting the face image according to the content of the foreign object for the deleted part. For example, when the foreign object is an eye patch, the face image correction unit extracts an image of the eye on the side that is not hidden by the eye patch and horizontally inverts the eye image and the eye patch part obtained by the left and right inversion. The face image is corrected by combining the deleted face image. The face image determination unit performs processing for determining who the person to be recognized is based on the corrected face image.
JP 2003-132339 A

  However, in the technique of Patent Document 1, when there are a plurality of persons recorded in the photographed image from which the face image is extracted, each person's face image can be recognized, but the face image at which position corresponds to which person's face image. It cannot be specified whether it is a thing. The present invention has been made in view of such problems, and an object of the present invention is to provide a face detection apparatus, method, and program capable of identifying a plurality of face images in a captured image.

  In order to solve the above-described problems, a face detection device according to the present invention includes a captured image input unit that inputs a captured image in which a person's face portion is recorded, a detection unit that detects a person's face portion, and a memory stored in an IC card. A reading unit that reads the unique identification information, and a link unit that associates the detected face portion with the unique identification information.

  According to the present invention, the unique identification information stored in the IC card is associated with the detected face portion. That is, the captured image and the face portion can be quickly identified based on the unique identification information stored in the IC card.

  The face detection apparatus may further include a designation unit that receives designation of unique identification information, and an image processing unit that performs predetermined image processing on a face portion of a captured image corresponding to the designated unique identification information.

  The predetermined image processing is not particularly limited, such as color change and image size change. In this way, it is possible to provide an interesting image obtained by performing special image processing on the face portion corresponding to the specified unique identification information.

  The face detection apparatus further includes a template image input unit that inputs a template image having a composite area, which is a blank area in which the face part is arranged, and the image processing unit includes a face part corresponding to the specified unique identification information. An extraction unit that extracts a face image, which is an image, from a photographed image, and the extracted face image is arranged in a template image synthesis region, and a synthesized image is created by synthesizing the face image arranged in the synthesis region and the template image. You may have a synthetic | combination part.

  In this way, it is possible to provide an interesting image in which the face portion corresponding to the unique identification information is arranged in the synthesis area and synthesized.

  The face detection apparatus may further include an image processing unit that performs predetermined image processing on a face portion other than the face portion corresponding to the specified unique identification information.

  The predetermined image processing is not particularly limited, such as color change or image size change, but as an example is image processing that makes it difficult to view a face portion other than the face portion corresponding to the specified unique identification information.

  In this way, it is possible to prevent viewing and copying of face images other than the face portion corresponding to the specified unique identification information, which is useful for portrait rights protection and copyright protection.

  The link unit may associate the face part with the unique identification information according to the area on the captured image to which the face part belongs. Alternatively, the link unit may associate the face portion with the unique identification information according to the order in which the unique identification information is read.

  In order to solve the above-described problem, a face detection method according to the present invention includes a photographed image input step for inputting a photographed image in which a person's face is recorded, a detection step for detecting a person's face, and an IC card. A reading step for reading the unique identification information stored in the information processing method, and a link step for associating the detected face portion with the unique identification information.

  This face detection method has the same operational effects as the face detection device described above.

  In order to solve the above-described problem, a face detection program according to the present invention includes a photographed image input step for inputting a photographed image in which a person's face part is recorded, a detection step for detecting a person's face part, and an IC card. The computer is caused to execute a reading step for reading the unique identification information stored in the computer and a linking step for associating the detected face portion with the unique identification information.

  This face detection program has the same effects as the above-described face detection apparatus. The face detection program may be provided by being recorded on a CD-ROM, DVD, MO, or other computer-readable recording medium.

  According to the present invention, the unique identification information stored in the IC card is associated with the detected face portion. That is, the captured image and the face portion can be quickly identified based on the unique identification information stored in the IC card.

  Hereinafter, preferred embodiments of the present invention will be described with reference to the accompanying drawings.

<First Embodiment>
[Schematic configuration]
FIG. 1 is a schematic functional configuration diagram of a face detection apparatus 100 according to a preferred first embodiment of the present invention. The face detection device 100 includes information from a processing unit 1 constituted by a CPU, a one-chip microcomputer, a media reader 2 that reads a captured image from various recording media such as a compact flash and smart media, and an IC chip embedded in an IC card 6. IC card reader 3, management DB 4 composed of various mass storage devices such as a hard disk, and storage unit 5 composed of semiconductor memory. The processing unit 1 reads the captured image input unit 10, the face detection unit 11, the management number acquisition unit 12, and the link unit 13 that are programs stored in the storage unit 5 from the storage unit 5 as appropriate and executes them. The captured image input unit 10 inputs the captured image read from the media reader 2. The captured image input unit 10 may input a captured image input from a digital still camera (DSC) 7 or a film scanner via a USB port, a SCSI port, or the like, or FTP (File Transfer) from a network-connected personal computer. Protocol) A captured image may be input by a file upload function of a server.

  The face detection unit 11 detects a face portion of a person from a captured image stored in the storage unit 5 by a known face recognition technique. When a plurality of persons are recorded in the captured image, a plurality of face portions are individually detected. The face detection unit 11 specifies the face position, which is the position of the detected face part, for each face part and stores the face position in the storage unit 5.

  The management number acquisition unit 12 acquires a management number from information (including a start code, management number, end code, etc.) read from the IC card 6 by the IC card reader 3. This management number is unique identification information assigned to each IC card 6. The link unit 13 stores, in the management DB 4, face management information that associates the face position specified by the face detection unit 11 with the management number read from the IC card 6. A specific example of association will be described later.

[Process flow]
Next, the flow of link processing executed by the face detection apparatus 100 will be described based on the flowchart of FIG.

  In S1, the captured image input unit 10 inputs a captured image. FIG. 3 is an example of a captured image input to the captured image input unit 10. The photographed image is a group group photo in which a plurality of persons Fn (here, n = 1 to 4) are recorded. It is assumed that an IC card 6 on which a unique management number is recorded is distributed in advance to each person Fn.

  In S <b> 2, the face detection unit 11 detects a human face portion from the input captured image. In FIG. 3, the face portions f1 to f4 of the four persons F1 to F4 are detected. The face detection unit 11 specifies the face position, which is the position of the detected face part, for each face part and stores the face position in the storage unit 5. The face position is represented, for example, by the number of pixels from the origin O to the center position Cn of the face portion of the person Fn on a plane coordinate with the lower left corner of the photographed image as the origin O.

  In S3, the management number is acquired from the IC card 6. In S4, the management number acquired from the IC card 6 and the face position are associated with each other according to a predetermined rule and stored in the management DB 4.

  The rules for associating the unique management number recorded on the IC card 6 distributed to each person Fn with the face position Cn of each person Fn are not particularly limited, and can take various forms. For example, as shown in FIG. 4, a plurality of IC card readers 3-1 to 3-4 are installed for each photographing position assigned to a plurality of persons F1 to F4. Each person F1 to F4 is distributed with IC cards 6-1 to 3-4 in which a unique management number is recorded in advance. Each person F1 to F4 causes the IC card readers 3-1 to 3-4 installed at the photographing position where the person stands to read the information of the IC card 6 held by the person. After all of the persons F1 to F4 have the IC card readers 3-1 to 3-4 read the information, the DSC 7 records captured images (see FIG. 2) with the persons F1 to F4 as subjects. In step S <b> 1, the captured image recorded by the DSC 7 is input to the processing unit 1. In step S3, the card information in which the identification numbers 3-1 to 3-4 of the IC card readers 3-1 to 3-4 are associated with the management numbers read from the card readers 3-1 to 3-4 is obtained. Store in the storage unit 5. FIG. 5A shows an example of card information.

  On the other hand, as shown in FIG. 1, the storage unit 5 is previously overlapped with the identification numbers 3-1 to 3-4 of the IC card reader and the face positions specified from the subject standing at the installation position of the card reader. A face reference position table 5a is stored in which the face reference position, which is a non-performing area, is associated with the face reference position. FIG. 5B is an example of information stored in the face reference position table 5a. The face reference position is a closed region to which the face position of the person photographed at the installation position of the card reader 3 uniquely belongs. For example, as shown in FIG. 6, rectangular areas R1 to R4 that divide a captured image into four parts, that is, upper left, upper right, lower left, and lower right, can be used as the face reference position.

  The link unit 13 identifies the face reference position to which each face position belongs by referring to the face position identified from the photographed image in S2 and the face reference position table 5a, and identifies identification information (file name, etc.) of the photographed image and Face position information in which the face position detected from the captured image is associated with the face reference position to which the face position belongs is stored in the storage unit 5. FIG. 5C shows an example of face position information. Further, the link unit 13 associates the management number, the face position, and the identification information of the photographed image from which the face position is detected according to the card information, the face reference position table 5a, and the face position information, and manages this correspondence as face management information. Store in DB4. FIG. 5D shows an example of face management information.

  Alternatively, the correspondence between the unique management number recorded on the IC card 6 distributed to each person Fn and the face position of each person may be performed as follows. As shown in FIG. 7, the IC card reader 3 that sequentially reads the IC cards 6 distributed to a plurality of persons F1 to F4 is installed at a person passing position (such as an entrance gate of an attraction). Similarly to the above, each person F1 to F4 is distributed with an IC card 6 in which a unique management number is recorded in advance. Each person F1 to F4 causes the IC card reader 3 to read the information of the IC card 6 possessed. Thereafter, each of the persons F1 to F4 is guided to a position corresponding to the order in which the cards are read, and a photographed image in which the persons F1 to F4 at the positions are recorded is input in S1. In step S <b> 3, card information in which the management number acquired from the IC card 6 is associated with the order of reading each management number of the IC card 6 is stored in the storage unit 5. FIG. 8A shows an example of card information.

  On the other hand, in the storage unit 5 in advance, the order of reading each management number of the IC card 6 and the face reference position which is a non-overlapping area to which the face position specified from the subject standing at the position where each person is guided belongs. Is stored in the face reference position table 5a. FIG. 8B is an example of information stored in the face reference position table 5a. The link unit 13 compares the face position detected in S2 with the face reference position table 5a, specifies the face reference position to which each face position belongs, each face position, the face reference position to which the face position belongs, and the captured image. The face position information associated with the identification information is stored in the storage unit 5. FIG. 8C shows an example of face position information. Further, the link unit 13 associates the management number with the face position in accordance with the card information, the face reference position table 5a, and the face position information, and stores this correspondence in the management DB 4 as face management information. FIG. 8D shows an example of face management information. In this way, the management number and the face position can be associated with each other without installing the IC card reader 3 for each position where a person stands, and capital investment can be suppressed.

Second Embodiment
Various image processing can be performed on a specific face portion of the photographed image according to the face management information obtained by the face detection apparatus 100 of the first embodiment. For example, the face portion other than the face portion corresponding to the designated management number may be subjected to various protection processes such as mosaic processing, color change, brightness change, and shape change so as not to be browsed. FIG. 9 is a schematic functional configuration diagram of a captured image providing system that performs such image processing. The photographed image providing system shown in FIG. 9 is configured by connecting a print order server 200 and a personal computer 300 to each other via a network 400 such as the Internet. The print order server 200 is connected to the face detection apparatus 100 via a LAN or the like, and shares face management information in the management DB 4 of the face detection apparatus 100. In this figure, only one personal computer 300 is shown, but actually, a plurality of personal computers 300 are connected to the print order server 200. In this case, the print order server 200 identifies the personal computer 200 with unique identification information (IP address, user ID, etc.).

  The print order server 200 accepts an order such as selection of a desired photographed image, the size of a photographic print, the number of prints, and the like by the order unit 203 and prints it by the printer 204. Data (such as a web page) for displaying a preview of the photographed image ordered based on this is transmitted to the personal computer 300. In this preview display, only the face portion corresponding to the management number transmitted from the personal computer 300 to the print order server 200 is displayed, and other face portions are subjected to various image protection processing such as mosaic processing so that they cannot be viewed. Is done. This image processing is performed by the image processing unit 14 of the face detection apparatus 100.

  FIG. 10 is a flowchart of image providing processing executed by the captured image providing system. The orderer who has the IC card 6 inputs a management number from the personal computer 300 to the personal computer 300 with the IC card reader 3 when ordering photo prints, or the orderer operates the keyboard or the like on the management number printed on the IC card 6. Manually input from the unit 301 to the personal computer 300 and transmit the management number to the print order server 200 (S101). The order unit 203 of the print order server 200 notifies the image processing unit 14 of the received management number and requests image protection processing for the face portion corresponding to the management number (S102). The image processing unit 14 receives the designation of the management number by this notification, and specifies the captured image and the face portion corresponding to the designated management number according to the face management information. The captured image corresponding to the management number transmitted from the personal computer 300 can be quickly identified by the face management information. The image processing unit 14 processes the specified captured image so that it is difficult or impossible to browse a face portion other than the specified face portion (mosaic processing or the like) (S103). The captured image after the image processing unit 14 has processed is referred to as a processed image. In addition, processing that makes it difficult to view all the face portions in advance is performed on the captured image, and processing (such as mosaic cancellation) is performed so that only the face portion corresponding to the designated management number can be viewed. Good.

  In FIGS. 11A to 11D, it is difficult to view the face portions other than the face portion fn of the face positions (xn, yn) to (xn, yn) corresponding to the management numbers n (n = 1 to 4), respectively. It is an example of the process image processed so that it may become.

  The image processing unit 14 transmits the processed image to the print order server 200 (S104). The browsing unit 202 of the print order server configured with the Web server transmits a Web page for browsing the processed image to the personal computer 300 (S105). The personal computer 300 displays the received web page with a web browser or the like (S106). However, in this Web page, it is difficult to view a face portion other than the face portion corresponding to the designated management number. In other words, face images other than the face portion corresponding to the management number recorded on the IC card 6 possessed by each person can be prevented from being freely viewed and duplicated on the personal computer 300, thereby protecting portrait rights and copyright protection. Useful.

<Third Embodiment>
Image processing executed by the image processing unit 14 is not limited to image protection processing. For example, the detected face portion may be extracted, and a composite image may be created by combining the face portion with a blank area of a “crisp” template image in which the face portion of the human character is cut out. At this time, the face image corresponding to the designated management number may be arranged in a specific composition area.

  FIG. 12 is a schematic functional configuration diagram of the face detection apparatus 100 that executes such image processing. The face detection apparatus 100 further includes a template input unit 15. The template input unit 15 includes a media reader, a USB port, and the like, and receives template image input. As shown in FIG. 13, the template image has a synthesis area (Pn: n = 1 to 4 in this figure) that is a “clear” -like blank area that synthesizes the face portion. The shape and number of composite areas and the design of the template image are not limited to those shown in the figure.

  The image processing unit 14 includes a face extraction unit 14a and a synthesis unit 14b. The face extraction unit 14a extracts a face image that is an image of the face portion detected by the face detection unit 11 from the captured image. The extracted face image is stored in the storage unit 5. The synthesizing unit 14b creates a synthesized image by arranging the face image in the storage unit 5 in the template image synthesizing region. The created composite image may be sent to the personal computer 300 via the network 400 by the browsing unit 202 of the print order server 200 or printed by the printer 204 as in the second embodiment. Good.

  The face detection apparatus 100 according to the present embodiment performs the same processing as in FIG. 10 except for the following points. That is, in S103, the compositing unit 14b arranges an image of the face portion corresponding to the management number notified from the print order server 200 among the face images extracted by the face extracting unit 14a in a specific composite region. Perform predetermined image processing (such as changing the length, shape, and color of the face image according to the composition area) on the face part image corresponding to the notified management number, and place it in a specific composition area Also good. The specific synthesis area is, for example, the most conspicuous position in the design of the template image, but is not limited to this.

  FIG. 14 is an example in which the face portion corresponding to the designated management number is arranged at a specific composition position. Here, the synthesizing unit 14b shows a state where the face portion f1 corresponding to the management number 001 transmitted / designated from the personal computer 300 is synthesized at a specific synthesis position, that is, the most prominent synthesis position P1. The other face portions f2 to f4 are appropriately arranged, for example, by being randomly arranged at the remaining synthesis positions P2 to P4. The combining unit 14b combines the face portion arranged at each combining position with the template image, and transmits the obtained combined image as a processed image to the print order server 200 (S105). Since the processing of other steps is the same as that of the second embodiment, description thereof is omitted.

  By doing so, it is possible to provide an interesting image obtained by performing special image processing on the face portion corresponding to the management number of the IC card 6 held by the orderer among the face images of the person extracted from the group photo. Can do. In particular, the face portion corresponding to the designated management number can be arranged in a specific composition area of the template image.

Schematic functional block diagram of the face detection apparatus according to the first embodiment Flow chart showing the flow of link processing The figure which shows an example of the photographed image The figure which shows an example of the IC card reader arrange | positioned in the standing position of a person Conceptual explanatory diagram of information stored in storage unit and management DB A diagram exemplifying the divided areas on the captured image The figure which illustrated the situation which guides the person who has an IC card to the position according to the order which made the IC card read Conceptual explanatory diagram of information stored in storage unit and management DB Schematic functional configuration diagram of a captured image providing system according to the second embodiment Flowchart of image providing process executed by photographed image providing system The figure which shows an example of the picked-up image processed so that face parts other than the face part corresponding to a management number may become difficult to browse Schematic functional configuration diagram of a face detection apparatus according to the third embodiment The figure which shows an example of a template image Diagram showing an example of a composite image

Explanation of symbols

1: processing unit, 2: media reader, 3: IC card reader, 4: management DB, 10: captured image input unit, 11: face detection unit, 12: management number acquisition unit, 13: link unit, 100: face detection apparatus

Claims (10)

A captured image input unit for inputting a captured image in which a person's face is recorded;
A detection unit for detecting a face portion of the person;
A reading unit for reading unique identification information stored in the IC card;
A link unit that associates the detected face part with the unique identification information;
A face detection device comprising:   The face detection apparatus according to claim 1, further comprising a designation unit that accepts designation of unique identification information.   The face detection apparatus according to claim 2, further comprising an image processing unit that performs predetermined image processing on a face portion of a captured image corresponding to the specified unique identification information. A template image input unit for inputting a template image having a composite area that is a blank area in which the face portion is arranged;
The image processing unit includes: an extraction unit that extracts a face image that is an image of a face part corresponding to the specified unique identification information from the photographed image; and the extracted face image as a combined region of the template image. The face detection device according to claim 3, further comprising: a combining unit configured to generate a combined image by combining the face image disposed in the combining region and the template image.   The face detection apparatus according to claim 2, further comprising an image processing unit that performs predetermined image processing on a face part other than the face part corresponding to the specified unique identification information.   The face detection apparatus according to claim 5, wherein the predetermined image processing is image processing that makes it difficult to view a face portion other than the face portion corresponding to the specified unique identification information.   The face detection device according to claim 1, wherein the link unit associates the face part with the unique identification information according to a region on a captured image to which the face part belongs.   The face detection device according to claim 1, wherein the link unit associates the face portion with the unique identification information according to an order in which the unique identification information is read. A captured image input step for inputting a captured image in which a person's face is recorded;
A detection step of detecting a face portion of the person;
A reading step of reading unique identification information stored in the IC card;
A link step of associating the detected face part with the unique identification information;
A face detection method including: A captured image input step for inputting a captured image in which a person's face is recorded;
A detection step of detecting a face portion of the person;
A reading step of reading unique identification information stored in the IC card;
A link step of associating the detected face part with the unique identification information;
A face detection program that runs a computer.

Images