WO2017017900A1 - 顔照合装置およびこれを備えた顔照合システムならびに顔照合方法 - Google Patents
顔照合装置およびこれを備えた顔照合システムならびに顔照合方法 Download PDFInfo
- Publication number
- WO2017017900A1 WO2017017900A1 PCT/JP2016/003174 JP2016003174W WO2017017900A1 WO 2017017900 A1 WO2017017900 A1 WO 2017017900A1 JP 2016003174 W JP2016003174 W JP 2016003174W WO 2017017900 A1 WO2017017900 A1 WO 2017017900A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- registrant
- image
- face
- images
- collation
- Prior art date
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/94—Hardware or software architectures specially adapted for image or video understanding
- G06V10/945—User interactive design; Environments; Toolboxes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/40—Software arrangements specially adapted for pattern recognition, e.g. user interfaces or toolboxes therefor
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/172—Classification, e.g. identification
- G06V40/173—Classification, e.g. identification face re-identification, e.g. recognising unknown faces across different face tracks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/50—Maintenance of biometric data or enrolment thereof
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30196—Human being; Person
- G06T2207/30201—Face
Definitions
- the present disclosure relates to a face collation apparatus that collates a face image of a person to be collated captured by an imaging apparatus with a face image of a registrant registered in advance, a face collation system including the face collation system, and a face collation method.
- a face matching device that performs face matching of a person to perform person authentication or the like.
- This type of face matching device performs person authentication and the like by capturing a face image of a person to be verified with an imaging device and comparing the captured face image with a face image of a registrant registered in advance.
- Such a face collation apparatus can be used for a security system, a surveillance camera system, and the like for the purpose of security, crime prevention, security, shoplifting prevention, customer recognition, and the like (for example, see Patent Document 1).
- the face image (hereinafter referred to as “captured image”) of the person to be collated captured by the imaging device is a face image of the registrant registered in advance (hereinafter referred to as “registrant image”). If it did not match, it could not be used for the next verification. Even if the captured image is the same person as the person in the registrant image due to the face direction, light direction, lighting environment, etc. of the person to be collated, the collation may not match. If at least a part of the captured image whose collation does not match the registrant image can be added to the already registered registrant image, collation accuracy in the next and subsequent collations can be improved. In addition, if at least a part of a captured image whose collation does not match the registrant image can be newly registered as a collation face image, a database of collation images can be created.
- the present disclosure has been devised in view of the above-described problems of the prior art, and by registering at least a part of a captured image that does not match the registrant image as a face image for collation, the next and subsequent times. It is a main object of the present invention to provide a face collation apparatus, a face collation system and a face collation method including the face collation apparatus capable of improving the collation accuracy in the collation and creating a database of collation images.
- a face collation device is a face collation device that collates a face image of a person to be collated captured by an imaging apparatus with a face image of a registrant registered in advance, and the person to be collated captured by the imaging apparatus
- An image acquisition unit that acquires a face image of the registrant, a registrant image storage unit that stores a face image of a registrant registered in advance, and an image verification unit that matches the face image of the person to be verified with the face image of the registrant
- a non-registrant image storage unit that stores the face image of the person to be collated that does not match the face image of the registrant as a non-registrant image, and the non-registration stored in the non-registrant image storage unit
- An image output unit for outputting at least one of the person images to a display device, and the non-registrant image selected to be registered by a user among the non-registrant images output to the display device.
- At least a part of a captured image that does not match the registrant image is registered as a face image for collation, thereby improving collation accuracy in the next and subsequent collations or creating a database of collation images. Can be achieved.
- FIG. 1 is a configuration diagram illustrating an outline of a face collation apparatus system according to the present disclosure.
- FIG. 2 is a block diagram illustrating a hardware configuration for realizing the face matching device system of the present disclosure.
- FIG. 3 is a flowchart showing a flow of processing for registering a non-registrant image in the non-registrant image storage unit.
- FIG. 4 is a flowchart showing a flow of processing for registering a non-registrant image in the registrant image storage unit.
- FIG. 5 is a diagram illustrating an example of a non-registrant image group screen displayed on the display device.
- FIG. 6 is a diagram showing a state in which a message box for confirming that the user selects and registers a non-registrant image is displayed on the display screen.
- FIG. 7 is a diagram illustrating a state where a check box on the display screen is unchecked.
- FIG. 8 is a diagram showing a state in which an input box for inputting a registered name is displayed on the display screen.
- FIG. 9 is a diagram showing a state in which a message box for confirming registration to the user is displayed on the display screen.
- a first disclosure made in order to solve the above-described problem is a face matching device for matching a face image of a person to be collated captured by an imaging device with a face image of a registrant registered in advance.
- Acquires the face image of the person to be collated captured by the imaging device stores the face image of the registrant registered in advance, collates the face image of the person to be collated with the face image of the registrant,
- the face image of the person to be collated that does not match the face image is stored as a non-registrant image, and at least one of the non-registrant images stored in the non-registrant image storage unit is output to the display device and displayed.
- the non-registrant image selected to be registered by the user is registered in the registrant image storage unit as a face image for verification.
- the captured image that did not match the registrant image is displayed on the display device as a non-registrant image, and the user among the non-registrant images displayed on the display device Since the non-registrant image selected to be registered by the registration is registered as a registrant image, at least a part of the captured image that does not match the registrant image can be registered as a face image for verification. It becomes possible. As a result, it is possible to improve collation accuracy in the subsequent collation or to create a database of collation images.
- the face collation device outputs the non-registrant image selected by the user to be registered among the non-registrant images output to the display device. It is registered as an image or registered as a face image of a new registrant.
- the captured image that does not match the registrant image is registered as a face image of an existing registrant already registered in the registrant image storage unit, or Since it is configured to register as a face image of a new registrant, if the collation target person is the same person as the existing registrant, the captured image of the collation target person is registered as the face image of the existing registrant. be able to. As a result, the number of face images of existing registrants can be increased, so that it is possible to improve collation accuracy in the next and subsequent collations.
- the captured image of the person to be collated is registered as a face image of the new registrant.
- the face collation apparatus groups the stored non-registrant images based on the similarity between the non-registrant images to create a non-registrant image group. It outputs to a display apparatus for every non-registrant image group.
- the face collation device since the non-registrant images are grouped based on the similarity between the non-registrant images and output to the display device for each group, the user can A plurality of non-registrant images similar to each other can be confirmed or compared together. Thereby, the user can easily perform an operation of selecting whether or not to register a non-registrant image as a registrant image.
- the fourth disclosure is characterized in that the non-registrant image groups are output in descending order of the number of non-registrant image contents.
- the non-registrant image groups are configured to be output in descending order of the number of non-registrant images, the user can use the non-registrant images as registrant images.
- the operation of selecting whether or not to register can be performed in order from a non-registrant image group having a large number of non-registrant images.
- a non-registrant image group with a high content of non-registrant images is a group that includes captured images with a high possibility of collation failure or a person with a high appearance frequency.
- the fifth disclosure is characterized in that only non-registrant image groups in which the number of non-registrant images included exceeds a predetermined number are output to the display device.
- the configuration is such that only the registrant image groups whose number of non-registrant images exceeds the predetermined number is output to the display device, Registrant image groups whose image content is equal to or less than a predetermined number are not output to the display device. This eliminates the need for registration of non-registrant images for registrant image groups whose number of non-registrant images is equal to or less than a predetermined number, so that registration of non-registrant images can be performed efficiently. Is possible.
- the sixth disclosure is characterized in that the re-learning of the verification data is performed using the non-registrant image registered in the registrant image storage unit.
- the image collation unit is configured to relearn the collation data using the non-registrant image newly registered in the registrant image storage unit.
- the image collation unit is configured to relearn the collation data using the non-registrant image newly registered in the registrant image storage unit.
- a seventh disclosure relates to the face matching device, an imaging device that captures a face image of a person to be verified, a display device that displays a non-registrant image output from the face matching device, and a non-registrant image It is a face collation system provided with the input device for a user to input the command regarding registration into a face collation apparatus.
- FIG. 1 is a configuration diagram illustrating an outline of a face matching system according to the present disclosure.
- This face collation system collates a face image (hereinafter referred to as “captured image”) of a person to be collated captured by an imaging device with a face image (hereinafter referred to as “registrant image”) registered in advance.
- This is a system for performing person authentication, and can be applied to a security system, a surveillance camera system, and the like.
- the face matching system 1 is applied to a security system, it is used in combination with an entrance management device that unlocks an electric key or an automatic door.
- the face matching system 1 is applied to a surveillance camera system, it is used in combination with a recording device or an alarm device.
- the face matching system 1 includes an imaging device 2, a face matching device 3, a display device 4, and an input device 5.
- the face matching device 3 is connected to the imaging device 2, the display device 4, and the input device 5 via wired or wireless communication such as a LAN (Local Area Network).
- the face matching device 3 and the imaging device 2, the connection form with the display device 4 and the input device 5 is not particularly limited, and may be, for example, a connection through a network such as the Internet, or a general-purpose interface (for example, USB (Universal Serial Bus)) May be connected to each other via a communication cable.
- a network such as the Internet
- a general-purpose interface for example, USB (Universal Serial Bus)
- the imaging device 2 is, for example, a CCD camera, and is installed on a wall or ceiling, and captures a face image of the person to be verified. As long as at least the face image of the person to be collated can be captured, the imaging device 2 is not particularly limited in its form, function, arrangement, quantity, and the like, and various changes can be made.
- the imaging device 2 may be a known omnidirectional network camera.
- the face image (captured image) of the person to be collated captured by the image capturing device 2 is input to the face collating device 3.
- the display device 4 is a general display device (monitor device), and displays the non-registrant image output from the face matching device 3. Details of the non-registrant image will be described later.
- the input device 5 is a general input device such as a keyboard or a mouse, and is used by a user to input a command for registering a non-registrant image displayed on the display device 4 to the face matching device 3.
- the display device 4 and the input device 5 are not particularly limited in terms of form, function, arrangement, quantity, and the like as long as non-registrant image display and non-registrant image registration instructions can be input. Instead, various modifications are possible.
- the face matching device 3 is a general computer device. Although not shown in detail, a CPU (Central Processing Unit) that centrally executes various types of information processing and control of peripheral devices based on a predetermined control program, the CPU's It includes RAM (Random Access Memory) that functions as a work area, ROM (Read Only Memory) that stores control programs and data executed by the CPU, and a network interface that executes communication processing via the network. .
- Various functions (for example, image matching processing) of the face matching device 3 described in detail later can be realized by the CPU executing a predetermined control program (for example, an image matching program).
- the face matching device 3 is not limited to a computer device, and other information processing devices (such as a server) that can perform the same function can also be used. Further, at least a part of the functions of the face matching device 3 may be replaced by processing using other known hardware.
- the face collation apparatus 3 includes an image acquisition unit 11, a registrant image storage unit 12, an image collation unit 13, a non-registrant image storage unit 14, an image group creation unit 15, and an image An output unit 16 and an image registration unit 17 are provided. Each unit is controlled by a control unit (not shown).
- the image acquisition unit 11 is connected to the imaging device 2 and acquires the face image (captured image) of the person to be verified from the imaging device 2.
- the image acquisition unit 11 is connected to the image verification unit 13, and the captured image acquired by the image acquisition unit 11 from the imaging device 2 is input to the image verification unit 13.
- the registrant image storage unit 12 stores a registered registrant's face image (registrant image).
- One registrant image is a face image of one person.
- the registrant image storage unit 12 generally stores a plurality of registrant images.
- the registrant image storage unit 12 may store one registrant image for each of a plurality of persons, or may store a plurality of registrant images for each of a plurality of persons.
- the registrant image is given a registrant name as a name for identifying the subject.
- the registrant name may be given by the user, or the registrant image storage unit 12 may automatically give it according to a predetermined method.
- the image collation unit 13 is connected to the image acquisition unit 11, the registrant image storage unit 12, and the non-registrant image storage unit 14.
- the image collation unit 13 collates the captured image input from the image acquisition unit 11 with the registrant image stored in the registrant image storage unit 12. Specifically, when the similarity between the captured image and the registrant image is calculated using a model obtained by machine learning performed in advance, and the calculated similarity exceeds a preset threshold value Determines that the captured image matches the registrant image, and determines that the captured image does not match the registrant image if it is equal to or less than the threshold.
- the above-described machine learning is performed using data including feature amounts used for collation of face images, which is referred to as collation data.
- the face image matching technique used in the image matching unit 13 is not particularly limited, and various other known face image matching techniques can be used.
- the collation result (that is, the authentication result) in the image collation unit 13 is output to the outside of the face collation device 3 such as a security system or a surveillance camera system via an output unit (not shown).
- the captured image that does not match the registrant image in the collation by the image collation unit 13 is input to the non-registrant image storage unit 14.
- the captured image that matches the registrant image in the collation by the image collation unit 13 is input to the registrant image storage unit 12 and registered as a registrant's face image already registered in the registrant image storage unit 12. Good.
- the non-registrant image storage unit 14 stores a captured image that is input from the image collating unit 13 and does not match the registrant image.
- the captured image stored in the non-registrant image storage unit 14 is referred to as a “non-registrant image”. Note that even if the captured image is the same person as the registrant who has already been registered in the registrant image storage unit 12 due to the face direction, light direction, lighting environment, etc. of the person to be collated, collation is possible. It may not match. Therefore, the non-registrant image may include a registrant's face image already registered in the registrant image storage unit 12.
- the image group creation unit 15 is connected to the non-registrant image storage unit 14 and groups the non-registrant images stored in the non-registrant image storage unit 14 based on the similarity between the non-registrant images. To create a non-registrant image group. Specifically, the similarity between all the non-registrant images recorded in the non-registrant image storage unit 14 is calculated. The similarity can be obtained as a sum of absolute values or a sum of squares of differences between pixel values of pixels corresponding to each other between two images. Or after projecting a non-registrant image on the feature space used for collation, you may obtain
- the grouping technique used in the image group creation unit 15 is not particularly limited, and various other known grouping techniques can be used.
- the image output unit 16 is connected to the non-registrant image storage unit 14 and the display device 4, and displays the non-registrant image stored in the non-registrant image storage unit 14 for each non-registrant image group. Output to. Thereby, the non-registrant image group is displayed for each group on the screen of the display device 4. For example, the same group is displayed on the same screen of the display device 4.
- the non-registrant image group may be displayed on the display device 4 in descending order of the content of the non-registrant image.
- a non-registrant image group with a high content of non-registrant images is a group that includes captured images with a high possibility of collation failure or a person with a high appearance frequency.
- the image output unit 16 may display only the non-registrant image groups in which the number of non-registrant images included exceeds a predetermined number on the display device 4. That is, non-registrant image groups whose number of non-registrant images is not more than a predetermined number may not be output to the display device 4. In this way, the registration work for the non-registrant image is not necessary for the registrant image group in which the number of non-registrant image content is equal to or less than the predetermined number. Can be done.
- the method for outputting the non-registrant image to the display device 4, that is, the method for displaying the non-registrant image on the display device 4, is not limited to this.
- one or a plurality of non-registrant images that are not grouped may be output to the display device 4 instead of each non-registrant image group.
- the image registration unit 17 is connected to the input device 5, and a user command is input from the input device 5.
- the user command is a command related to registration of a non-registrant image, including a command to select a non-registrant image to be registered and a command to register the selected non-registrant image. Details will be described later with reference to FIGS.
- the image registration unit 17 is connected to the non-registrant image storage unit 14 and the registrant image storage unit 12, and is selected by the user from the non-registrant images output to the display device 4.
- the non-registrant image is registered in the registrant image storage unit 12.
- the non-registrant image selected to be registered by the user may be registered as a face image of a registrant already registered in the registrant image storage unit 12, or registered as a face image of a new registrant. May be.
- FIG. 2 is a block diagram showing a hardware configuration for realizing the face matching system 1 of FIG.
- the face matching system 1 in FIG. 1 includes a photographing unit 21, a display unit 22, an input unit 23, a communication unit 24, a storage unit 25, a processor 26, and a system bus 27 for connecting them. It is realized by the hardware configuration including.
- the imaging unit 21 is, for example, a CCD camera, and is used to capture a face image of the person to be verified.
- the imaging unit 21 corresponds to the imaging device 2 in FIG.
- the display unit 22 is a general display device, and is used to display the non-registrant image output from the face matching device 3.
- the display unit 22 corresponds to the display device 4 in FIG.
- the input unit 23 is a general input device such as a keyboard or a mouse, and is used by the user to input a command to the face matching device 3.
- the input unit 23 corresponds to the input device 5 in FIG.
- the communication unit 24 is a general communication module, and is used for the face matching device 3 to communicate with other external devices or systems.
- the communication unit 24 is used, for example, to output the collation result in the image collation unit 13 to the display unit 22, an external security system, a surveillance camera system, or the like.
- the storage unit 25 is, for example, a ROM or a hard disk, and stores various programs and various data for realizing various functions of the face matching device 3.
- the various programs are the image acquisition unit 11, the registrant image storage unit 12, the image collation unit 13, the non-registrant image storage unit 14, the image group creation unit 15, the image output unit 16, and the image registration unit 17 of the face collation apparatus 3.
- the various types of data include the registrant image stored in the registrant image storage unit 12, the non-registrant image and non-registrant image group stored in the non-registrant image storage unit 14, and the above-described units of the face matching device 3.
- the processor 26 is, for example, a CPU, reads the above-described various programs and various data from the storage unit 25 onto a RAM (not shown), and acquires the image acquisition unit 11, the image verification unit 13, the image group creation unit 15 of the face verification device 3, Each process in the image output unit 16 and the image registration unit 17 is executed. In addition, the processor 26 performs overall control of the face matching device 3 and control of peripheral devices (the imaging device 2, the display device 4, and the input device 5) of the face matching device 3.
- FIG. 3 is a flowchart showing a flow of processing for registering a non-registrant image in the non-registrant image storage unit 14
- FIG. 4 is a flowchart of processing for registering a non-registrant image in the registrant image storage unit 12. It is a flow to show.
- the imaging device 2 captures the face image of the person to be verified (step ST101).
- the image acquisition unit 11 of the face matching device 3 acquires a face image (captured image) of the person to be verified from the imaging device 2 (step ST102).
- the image collation unit 13 of the face collation apparatus 3 collates the captured image with the registrant image stored in the registrant image storage unit 12, and determines whether or not they match (step ST103).
- step ST104 the captured image is registered in the registrant image storage unit 12, and in step ST105, the captured image is registered in the non-registrant image storage unit 14. In this way, the captured image that does not match the registrant image in the collation by the image collation unit 13 is registered in the non-registrant image storage unit 14 as a non-registrant image.
- the image group creation unit 15 of the face collation apparatus 3 groups the non-registrant images stored in the non-registrant image storage unit 14 based on the similarity between the non-registrant images, and the non-registrant image group. Is created (step ST201). Subsequently, the non-registrant image group is displayed on the display device 4 for each group (step ST202). At this time, it is preferable to output the non-registrant image groups in descending order of the number of non-registrant images.
- FIG. 5 is a diagram showing an example of a non-registrant image group screen (hereinafter simply referred to as “display screen”) 31 displayed on the display device 4.
- the display screen 31 shown in FIG. A non-registrant image group 33 including a single non-registrant image 32 is displayed, and a user selects a non-registrant image 32 to be registered in the registrant image storage unit 12 on the right side of each non-registrant image 32. 5, the check boxes 34 of all the non-registrant images 32 are checked and are not registered in the registrant image storage unit 12 by removing the check. The non-registrant image 32 is selected.
- a display box 35 for displaying the name of the non-registrant image group 33 and the number of images of the non-registrant image 32 included in the non-registrant image group 33 is provided at the upper left of the display screen 31.
- the name of the non-registered person image group 33 is “non-registered person 1”, and the number of non-registered person images 32 is “24”.
- a button 36 for switching the display screen 31 for displaying the next non-registrant image group is provided on the upper right of the display screen 31.
- step ST203 it is confirmed to the user whether or not a registration operation for registering the non-registrant image 32 in the registrant image storage unit 12 is performed.
- a message box 41 “Do you want to select and register an image?” Is displayed on the display screen 31 to confirm with the user.
- the user selects whether or not to perform registration work. The user can select whether to perform the registration work by operating the input device 5 and selecting either “Yes” or “No” in the message box 41. If the user selects to perform the registration work (ST204: Yes), the process proceeds to step ST205. If the user selects not to perform the registration work (ST204: No), the process proceeds to step ST207 described later.
- step ST205 the user selects a non-registrant image 32 to be registered in the registrant image storage unit 12.
- the non-registered person image group 33 may include face images of a person different from the person to be registered. Therefore, when registering the non-registrant image 32 in the registrant image storage unit 12, it is necessary to select and exclude the non-registrant image 32 that is not registered. Specifically, as shown in FIG. 7, by unchecking the check box 34 of the non-registrant image 32, the non-registrant image 32 that is not registered can be selected and excluded. In the display screen 31 shown in FIG.
- the non-registrant image 32 selected in step ST205 is named and registered in the registrant image storage unit 12.
- an input box 44 for inputting a registered name is displayed on the display screen 31.
- the registered name and representative image of the registrant image group already registered in the registrant image storage unit 12 are displayed on the lower side of the display screen 31.
- two registrant images 45 and 46 are displayed.
- the registration name of the registrant image 45 is “registrant 1”
- the registration name of the registrant image 46 is “registrant 2”.
- the user operates the input device 5 to input a registered name in the input box 44.
- the non-registrant image 32 is registered as a registrant's face image with that registered name.
- the non-registrant image 32 is registered as a face image of the new registrant.
- a message box 47 “Do you want to register?” Is displayed on the display screen 31 to confirm with the user whether or not to register. You may make it do. In this case, the user can select whether to register by operating the input device 5 and selecting either “OK” or “Cancel” in the message box 47.
- next step ST207 the user selects whether or not to continue the registration work for the next non-registrant image group. For example, a message box for confirming with the user whether or not to continue the registration process may be displayed after step ST206 is completed. Alternatively, after the end of step ST206, the screen may automatically transition to the display screen for the next non-registrant image group, and the user may select whether or not to continue the registration work on that display screen.
- ST207: Yes the process returns to step ST202, and the processes after step ST202 are repeated.
- the process proceeds to step ST208.
- step ST208 the image collation unit 13 re-learns the collation data using the non-registrant image registered in the registrant image storage unit 12.
- the non-registrant image registered in the registrant image storage unit 12 is an image confirmed by the user to be related to a specific registrant. Therefore, by performing relearning using such an image, the accuracy of the verification data is improved. Thereafter, the process ends.
- the non-registrant image selected to be registered by the user among the non-registrant images registered in the non-registrant image storage unit 14 is registered in the registrant image storage unit 12 as a registrant image.
- the non-registrant image may be registered as a registrant face image already registered in the registrant image storage unit 12, or may be registered as a new registrant face image.
- the image collation part 13 can aim at the improvement of collation precision in the collation after the next by re-learning the data for collation using the non-registrant image registered in the registrant image memory
- a face matching device and a face matching system and a face matching method including the face matching device according to the present disclosure register at least a part of a captured image that does not match a face image registered in advance as a face image for matching
- the present invention is useful as a face matching device, a face matching system and a face matching method that can improve the matching accuracy in the next and subsequent matching, or to create a database of images for matching.
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Human Computer Interaction (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Databases & Information Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Collating Specific Patterns (AREA)
Abstract
Description
2 撮像装置
3 顔照合装置
4 表示装置
5 入力装置
11 画像取得部
12 登録者画像記憶部
13 画像照合部
14 非登録者画像記憶部
15 画像グループ作成部
16 画像出力部
17 画像登録部
Claims (18)
- 撮像装置により撮像された照合対象者の顔画像を予め登録された登録者の顔画像と照合する顔照合装置であって、
前記顔照合装置は、
前記撮像装置により撮像された照合対象者の顔画像を取得し、
予め登録された登録者の顔画像を記憶し、
前記照合対象者の顔画像を前記登録者の顔画像と照合し、
前記登録者の顔画像と合致しなかった前記照合対象者の顔画像を非登録者画像として記憶し、
前記非登録者画像記憶部に記憶された前記非登録者画像のうちの少なくとも1つを表示装置に出力し、
前記表示装置に出力された前記非登録者画像のうちのユーザにより登録すべきと選択された前記非登録者画像を前記登録者画像記憶部に照合用の顔画像として登録する、
ことを特徴とする顔照合装置。 - 前記顔照合装置は、前記表示装置に出力された前記非登録者画像のうちの前記ユーザにより登録すべきと選択された前記非登録者画像を、既に登録されている登録者の顔画像として登録するか、または新規の登録者の顔画像として登録すること特徴とする請求項1に記載の顔照合装置。
- 前記顔照合装置は、記憶されている前記非登録者画像を前記非登録者画像間の類似度に基づいてグループ化して非登録者画像グループを作成し、
前記非登録者画像を前記非登録者画像グループ毎に前記表示装置に出力すること特徴とする請求項1に記載の顔照合装置。 - 前記顔照合装置は、前記非登録者画像グループを、前記非登録者画像の含有数が多い順に出力することを特徴とする請求項3に記載の顔照合装置。
- 前記顔照合装置は、前記非登録者画像の含有数が予め定められた数を超えた前記非登録者画像グループのみを前記表示装置に出力すること特徴とする請求項3に記載の顔照合装置。
- 前記顔照合装置は、前記登録者画像記憶部に登録された前記非登録者画像を用いて照合用データの再学習を行うことを特徴とする請求項1に記載の顔照合装置。
- 請求項1に記載の顔照合装置と、
照合対象者の顔画像を撮像する撮像装置と、
前記顔照合装置から出力された前記非登録者画像を表示する表示装置と、
前記非登録者画像の登録に関する命令をユーザが前記顔照合装置に入力するための入力装置と
を備えた顔照合システム。 - 前記顔照合装置は、前記表示装置に出力された前記非登録者画像のうちの前記ユーザにより登録すべきと選択された前記非登録者画像を、前記登録者画像記憶部に既に登録されている登録者の顔画像として登録するか、または新規の登録者の顔画像として登録すること特徴とする請求項7に記載の顔照合システム。
- 前記顔照合装置は、前記非登録者画像記憶部に記憶されている前記非登録者画像を前記非登録者画像間の類似度に基づいてグループ化して非登録者画像グループを作成し、
前記非登録者画像を前記非登録者画像グループ毎に前記表示装置に出力すること特徴とする請求項7に記載の顔照合システム。 - 前記顔照合装置は、前記非登録者画像グループを、前記非登録者画像の含有数が多い順に出力することを特徴とする請求項9に記載の顔照合システム。
- 前記顔照合装置は、前記非登録者画像の含有数が予め定められた数を超えた前記非登録者画像グループのみを前記表示装置に出力すること特徴とする請求項9に記載の顔照合システム。
- 前記顔照合装置は、前記非登録者画像を用いて照合用データの再学習を行うことを特徴とする請求項7に記載の顔照合システム。
- 撮像装置により撮像された照合対象者の顔画像を予め登録された登録者の顔画像と照合する顔照合装置方法であって、
前記撮像装置により撮像された照合対象者の顔画像を取得し、
予め登録された登録者の顔画像を記憶し、
前記照合対象者の顔画像を前記登録者の顔画像と照合し、
前記登録者の顔画像と合致しなかった前記照合対象者の顔画像を非登録者画像として記憶し、
前記非登録者画像のうちの少なくとも1つを表示装置に出力し、
前記表示装置に出力された前記非登録者画像のうちのユーザにより登録すべきと選択された前記非登録者画像を照合用の顔画像として登録する、
ことを特徴とする顔照合方法。 - 前記顔照合方法は、前記表示装置に出力された前記非登録者画像のうちの前記ユーザにより登録すべきと選択された前記非登録者画像を、既に登録されている登録者の顔画像として登録するか、または新規の登録者の顔画像として登録すること特徴とする請求項13に記載の顔照合方法。
- 前記顔照合方法は、前記非登録者画像を前記非登録者画像間の類似度に基づいてグループ化して非登録者画像グループを作成し、
前記非登録者画像を前記非登録者画像グループ毎に前記表示装置に出力すること特徴とする請求項13に記載の顔照合方法。 - 前記顔照合方法は、前記非登録者画像グループを、前記非登録者画像の含有数が多い順に出力することを特徴とする請求項15に記載の顔照合方法。
- 前記顔照合方法は、前記非登録者画像の含有数が予め定められた数を超えた前記非登録者画像グループのみを前記表示装置に出力すること特徴とする請求項15に記載の顔照合方法。
- 前記顔照合方法は、前記非登録者画像を用いて照合用データの再学習を行うことを特徴とする請求項13に記載の顔照合方法。
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2017530998A JP6865351B2 (ja) | 2015-07-27 | 2016-07-04 | 顔照合装置およびこれを備えた顔照合システムならびに顔照合方法 |
US15/527,160 US10558846B2 (en) | 2015-07-27 | 2016-07-04 | Face collation device, face collation system comprising same, and face collation method |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2015-147534 | 2015-07-27 | ||
JP2015147534 | 2015-07-27 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2017017900A1 true WO2017017900A1 (ja) | 2017-02-02 |
Family
ID=57884164
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2016/003174 WO2017017900A1 (ja) | 2015-07-27 | 2016-07-04 | 顔照合装置およびこれを備えた顔照合システムならびに顔照合方法 |
Country Status (3)
Country | Link |
---|---|
US (1) | US10558846B2 (ja) |
JP (1) | JP6865351B2 (ja) |
WO (1) | WO2017017900A1 (ja) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2018180666A1 (ja) * | 2017-03-30 | 2018-10-04 | ソニー株式会社 | 情報処理装置、情報処理方法及びプログラム |
JP2019028936A (ja) * | 2017-08-03 | 2019-02-21 | 日本電気株式会社 | 情報処理装置、人物同定情報決定方法、人物同定情報登録方法、およびプログラム |
JP2020022101A (ja) * | 2018-08-01 | 2020-02-06 | キヤノン株式会社 | 監視装置、生産ライン、および監視装置の制御方法 |
JP2021002084A (ja) * | 2019-06-19 | 2021-01-07 | 株式会社セブン銀行 | 認証システム、認証方法、および認証プログラム |
JP7415510B2 (ja) | 2019-12-09 | 2024-01-17 | 富士フイルムビジネスイノベーション株式会社 | 登録許可装置及び登録許可プログラム |
Families Citing this family (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP6700791B2 (ja) * | 2016-01-05 | 2020-05-27 | キヤノン株式会社 | 情報処理装置、情報処理方法及びプログラム |
CN109118616A (zh) * | 2018-07-09 | 2019-01-01 | 深圳市商汤科技有限公司 | 门禁控制方法及门禁控制装置 |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2011138387A (ja) * | 2009-12-28 | 2011-07-14 | Canon Inc | オブジェクト識別装置及びその制御方法 |
JP2012123676A (ja) * | 2010-12-09 | 2012-06-28 | Canon Inc | 個人認証装置 |
Family Cites Families (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7249320B2 (en) * | 2003-03-04 | 2007-07-24 | Microsoft Corporation | Method and system for displaying a title area for a page series |
WO2006049237A1 (ja) * | 2004-11-08 | 2006-05-11 | Matsushita Electric Industrial Co., Ltd. | 法線情報推定装置、登録画像群作成装置および画像照合装置ならびに法線情報推定方法 |
JP4140619B2 (ja) * | 2005-06-14 | 2008-08-27 | 松下電器産業株式会社 | 画像合成装置およびそれを用いた画像照合装置ならびに画像合成方法 |
JP4789658B2 (ja) * | 2006-03-14 | 2011-10-12 | 富士フイルム株式会社 | スライドショー再生方法、スライドショー再生プログラム、及びスライドショー再生装置 |
JP2007300363A (ja) * | 2006-04-28 | 2007-11-15 | Omron Corp | 監視装置および方法、並びにプログラム |
JP4264663B2 (ja) * | 2006-11-21 | 2009-05-20 | ソニー株式会社 | 撮影装置、画像処理装置、および、これらにおける画像処理方法ならびに当該方法をコンピュータに実行させるプログラム |
JP2008257487A (ja) | 2007-04-05 | 2008-10-23 | Multi Solution:Kk | 顔認証応用万引き検知システム |
JP5053731B2 (ja) * | 2007-07-03 | 2012-10-17 | キヤノン株式会社 | 画像表示制御装置及び画像表示制御方法及びプログラム及び記録媒体 |
JP2009271577A (ja) * | 2008-04-30 | 2009-11-19 | Panasonic Corp | 類似画像検索の結果表示装置及び類似画像検索の結果表示方法 |
JP2011081763A (ja) * | 2009-09-09 | 2011-04-21 | Sony Corp | 情報処理装置、情報処理方法及び情報処理プログラム |
JP4775515B1 (ja) * | 2011-03-14 | 2011-09-21 | オムロン株式会社 | 画像照合装置、画像処理システム、画像照合プログラム、コンピュータ読み取り可能な記録媒体、および画像照合方法 |
CN103196550A (zh) * | 2012-01-09 | 2013-07-10 | 西安智意能电子科技有限公司 | 一种对发射光源的成像信息进行筛选处理的方法与设备 |
WO2013114862A1 (ja) * | 2012-01-30 | 2013-08-08 | パナソニック株式会社 | 最適カメラ設定装置及び最適カメラ設定方法 |
JP5899472B2 (ja) * | 2012-05-23 | 2016-04-06 | パナソニックIpマネジメント株式会社 | 人物属性推定システム、及び学習用データ生成装置 |
-
2016
- 2016-07-04 JP JP2017530998A patent/JP6865351B2/ja active Active
- 2016-07-04 US US15/527,160 patent/US10558846B2/en active Active
- 2016-07-04 WO PCT/JP2016/003174 patent/WO2017017900A1/ja active Application Filing
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2011138387A (ja) * | 2009-12-28 | 2011-07-14 | Canon Inc | オブジェクト識別装置及びその制御方法 |
JP2012123676A (ja) * | 2010-12-09 | 2012-06-28 | Canon Inc | 個人認証装置 |
Cited By (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2018180666A1 (ja) * | 2017-03-30 | 2018-10-04 | ソニー株式会社 | 情報処理装置、情報処理方法及びプログラム |
JPWO2018180666A1 (ja) * | 2017-03-30 | 2020-02-06 | ソニー株式会社 | 情報処理装置、情報処理方法及びプログラム |
JP7070549B2 (ja) | 2017-03-30 | 2022-05-18 | ソニーグループ株式会社 | 情報処理装置、情報処理方法及びプログラム |
JP2022105553A (ja) * | 2017-03-30 | 2022-07-14 | ソニーグループ株式会社 | 情報処理装置、情報処理方法及びプログラム |
US11468592B2 (en) | 2017-03-30 | 2022-10-11 | Sony Corporation | Information processing apparatus and information processing method |
JP7400866B2 (ja) | 2017-03-30 | 2023-12-19 | ソニーグループ株式会社 | 情報処理装置、情報処理方法及びプログラム |
JP2019028936A (ja) * | 2017-08-03 | 2019-02-21 | 日本電気株式会社 | 情報処理装置、人物同定情報決定方法、人物同定情報登録方法、およびプログラム |
JP2020022101A (ja) * | 2018-08-01 | 2020-02-06 | キヤノン株式会社 | 監視装置、生産ライン、および監視装置の制御方法 |
JP2021002084A (ja) * | 2019-06-19 | 2021-01-07 | 株式会社セブン銀行 | 認証システム、認証方法、および認証プログラム |
JP7415510B2 (ja) | 2019-12-09 | 2024-01-17 | 富士フイルムビジネスイノベーション株式会社 | 登録許可装置及び登録許可プログラム |
Also Published As
Publication number | Publication date |
---|---|
US20170357845A1 (en) | 2017-12-14 |
JPWO2017017900A1 (ja) | 2018-05-17 |
JP6865351B2 (ja) | 2021-04-28 |
US10558846B2 (en) | 2020-02-11 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2017017900A1 (ja) | 顔照合装置およびこれを備えた顔照合システムならびに顔照合方法 | |
EP2980766A1 (en) | Monitoring device, monitoring system and monitoring method | |
US11682255B2 (en) | Face authentication apparatus | |
US20190080197A1 (en) | Communication terminal, communication system, and image processing method | |
JP2019125002A (ja) | 顔認証装置 | |
US8850299B2 (en) | Processing of biometric data by transformation | |
JP2006309491A (ja) | 生体認証システム | |
EP3594879A1 (en) | System and method for authenticating transactions from a mobile device | |
US20140211003A1 (en) | Safety monitoring device and method | |
JP6962431B2 (ja) | 顔認証装置 | |
CN113642639A (zh) | 活体检测方法、装置、设备和存储介质 | |
CN111768530A (zh) | 一种智能开锁方法及装置 | |
JP4830904B2 (ja) | ネットワークシステム | |
JP4943127B2 (ja) | 個人認証装置および個人認証システム | |
JP2019164834A (ja) | 顔認証装置 | |
JP2010176304A (ja) | 人物認証装置 | |
US11611552B2 (en) | Method for verifying an operator in a security system | |
JP2009129252A (ja) | 生体認証装置、生体認証方法およびコンピュータプログラム | |
WO2021005886A1 (ja) | 施設管理システム、顔認証装置、施設管理方法、及びコンピュータプログラム | |
JP6920647B2 (ja) | 集合住宅用インターホンシステム、集合住宅用インターホンシステムの制御方法、及びプログラム | |
JP2018074317A (ja) | 人物特定支援装置 | |
JP2018120301A (ja) | 顔認証システム、顔認証方法および顔認証プログラム | |
WO2020003400A1 (ja) | 顔照合システム | |
KR20180048043A (ko) | 스마트 작업지시 방법 | |
JPH0433065A (ja) | 指紋照合装置 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 16830008 Country of ref document: EP Kind code of ref document: A1 |
|
ENP | Entry into the national phase |
Ref document number: 2017530998 Country of ref document: JP Kind code of ref document: A |
|
WWE | Wipo information: entry into national phase |
Ref document number: 15527160 Country of ref document: US |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 16830008 Country of ref document: EP Kind code of ref document: A1 |