WO2021176945A1

WO2021176945A1 - Authentication device

Info

Publication number: WO2021176945A1
Application number: PCT/JP2021/004123
Authority: WO
Inventors: 亮野口; 一秀梅田; 伸介吉垣
Original assignee: Ｎｅｃソリューションイノベータ株式会社
Priority date: 2020-03-05
Filing date: 2021-02-04
Publication date: 2021-09-10
Also published as: JPWO2021176945A1; JP7456659B2

Abstract

An authentication device 400 includes: an acquiring unit 421 which acquires, from image data, a feature quantity of a part of a person in the image data, the person being identified by employing skeleton recognition to recognize the skeleton of the person; and an authenticating unit 422 which, on the basis of the feature quantity acquired by the acquiring unit 421, performs clothing authentication to determine whether the person in the image data is wearing clothing that has been registered in advance.

Description

Authentication device

The present invention relates to an authentication device, an authentication method, and a recording medium.

A technique called clothing authentication, which is an authentication process based on the feature amount of clothing in image data, is known.

For example, Patent Document 1 is a document that describes the authentication process based on the feature amount of clothes. According to Patent Document 1, the size of the clothing area is calculated based on the position information of the face area, and the direction in which the clothing area is arranged is determined based on the inclination information of the face. Then, clothing feature information is generated in an area based on the clothing area, and collation processing is performed. As described above, Patent Document 1 describes that the collation is performed based on the characteristic information of the clothing region estimated based on the face region.

JP-A-2010-219607

When collation based on the characteristic information of the clothing area calculated / estimated based on the face area as described in Patent Document 1, it is difficult to accurately calculate / estimate the clothing area depending on the state of the person or the like. There was something. As a result, there has been a problem that it is difficult to improve the accuracy of clothing authentication.

Therefore, an object of the present invention is to provide an authentication method, an authentication device, and a program that solve the problem that it is difficult to improve the accuracy of clothing authentication.

An authentication device, which is a form of the present disclosure, in order to achieve such an object
From the image data, an acquisition unit that acquires the feature amount of the part of the person specified by recognizing the skeleton of the person in the image data by skeleton recognition, and
Based on the acquired feature amount, an authentication unit that performs clothing authentication to determine whether or not the person in the image data is the pre-registered clothing, and an authentication unit.
It has a structure of having.

In addition, the authentication method, which is another form of the present disclosure,
Information processing device
From the image data, the feature amount of the part of the person specified by recognizing the skeleton of the person in the image data by skeleton recognition is acquired.
Based on the feature amount acquired by the acquisition unit, clothing authentication is performed to determine whether or not the person in the image data is the clothing registered in advance.

In addition, the recording medium which is another form of the present disclosure is
For information processing equipment
From the image data, an acquisition unit that acquires the feature amount of the part of the person specified by recognizing the skeleton of the person in the image data by skeleton recognition, and
Based on the feature amount acquired by the acquisition unit, an authentication unit that performs clothing authentication for determining whether or not a person in the image data is a pre-registered clothing, and an authentication unit.
It is a program to realize.

According to each configuration as described above, it is possible to provide an authentication device, an authentication method, and a recording medium capable of improving the accuracy of clothing authentication.

It is a figure which shows the configuration example of the authentication system in 1st Embodiment of this disclosure. It is a block diagram which shows the configuration example of the authentication apparatus shown in FIG. It is a figure which shows an example of the color information for each part for authentication shown in FIG. It is a figure which shows an example of the skeleton coordinate information shown in FIG. It is a figure which shows an example of the color information for each part shown in FIG. It is a figure for demonstrating the processing example of the skeleton recognition part and the part-specific color information acquisition part. It is a figure for demonstrating the processing example of the clothes authentication part. It is a flowchart which shows the operation example of the authentication apparatus in 1st Embodiment of this disclosure. It is a block diagram which shows the other configuration example of an authentication apparatus. It is a figure which shows an example of the hardware block diagram of the authentication apparatus in the 2nd Embodiment of this disclosure. It is a block diagram which shows the structural example of the authentication apparatus shown in FIG.

[First Embodiment]
The first embodiment of the present disclosure will be described with reference to FIGS. 1 to 9. FIG. 1 is a diagram showing a configuration example of the authentication system 100. FIG. 2 is a block diagram showing a configuration example of the authentication device 300. FIG. 3 is a diagram showing an example of color information 333 for each part for authentication. FIG. 4 is a diagram showing an example of skeleton coordinate information 335. FIG. 5 is a diagram showing an example of color information 336 for each part. FIG. 6 is a diagram for explaining a processing example of the skeleton recognition unit 343 and the color information acquisition unit 344 for each part. FIG. 7 is a diagram for explaining a processing example of the clothing authentication unit 345. FIG. 8 is a flowchart showing an operation example of the authentication device 300. FIG. 9 is a block diagram showing another configuration example of the authentication device 300.

In the first embodiment of the present disclosure, an authentication system 100 that performs clothing authentication for determining whether or not a person in image data is a pre-registered clothing will be described. As will be described later, the authentication system 100 acquires color information of a plurality of parts of a person specified based on the result of skeleton recognition for image data such as shoulders, elbows, and knees. Then, the authentication system 100 performs clothing authentication by comparing the acquired color information of a plurality of parts with the information stored in advance.

FIG. 1 shows a configuration example of the authentication system 100. Referring to FIG. 1, the authentication system 100 includes, for example, a camera 200 and an authentication device 300. As shown in FIG. 1, the camera 200 and the authentication device 300 are connected so as to be able to communicate with each other via, for example, a network.

The camera 200 is a photographing device that acquires image data. For example, the camera 200 is installed at a predetermined position, and acquires image data in which at least a person's face and upper body are reflected. Further, the camera 200 can transmit the acquired image data to the authentication device 300 in association with the acquisition date and time of the image data.

The authentication device 300 is an information processing device that authenticates the clothes worn by a person (clothes authentication). The authentication device 300 can also perform face authentication in addition to clothing authentication. FIG. 2 shows a configuration example of the authentication device 300. Referring to FIG. 2, the authentication device 300 has a screen display unit 310, a communication I / F unit 320, a storage unit 330, and an arithmetic processing unit 340 as main components.

The screen display unit 310 includes a screen display device such as an LCD (Liquid Crystal Display). The screen display unit 310 displays various information stored in the storage unit 330 such as image information 334, skeleton coordinate information 335, color information 336 for each part, and authentication result information 337 on the screen in response to an instruction from the arithmetic processing unit 340. Can be done.

The communication I / F unit 320 includes a data communication circuit. The communication I / F unit 320 performs data communication with a camera 200, an external device, or the like connected via a communication line.

The storage unit 330 is a storage device such as a hard disk or a memory. The storage unit 330 stores processing information and a program 338 required for various processes in the arithmetic processing unit 340. The program 338 realizes various processing units by being read and executed by the arithmetic processing unit 340. The program 338 is read in advance from an external device or a recording medium via a data input / output function such as the communication I / F unit 320, and is stored in the storage unit 330. The main information stored in the storage unit 330 includes, for example, facial feature amount information 331, learned model 332, authentication site-specific color information 333, image information 334, skeleton coordinate information 335, site-specific color information 336, and authentication. There is result information 337 and the like.

The face feature amount information 331 includes information indicating the face feature amount used when the face recognition unit 342 performs face recognition. In the face feature amount information 331, for example, identification information for identifying a person and information indicating the face feature amount are associated with each other. The face feature amount information 331 is acquired in advance from an external device or the like via, for example, the communication I / F unit 320 or the like, and is stored in the storage unit 330.

The trained model 332 is a trained model used when the skeleton recognition unit 343 performs skeleton recognition. The trained model 332 is generated in advance by performing machine learning using teacher data such as image data containing skeleton coordinates in an external device or the like, and is externally generated via a communication I / F unit 320 or the like. It is acquired from a device or the like and stored in the storage unit 330.

The trained model 332 may be updated by a re-learning process using additional teacher data.

The authentication part-specific color information 333 includes information indicating an object to be compared with the color information acquired from the image data when performing clothing authentication. For example, the color information 333 for each part for authentication is defined in advance. The color information 333 for each authentication part is acquired in advance from an external device or the like via, for example, the communication I / F unit 320 or the like, and is stored in the storage unit 330.

FIG. 3 shows an example of color information 333 for each part for authentication. Referring to FIG. 3, in the authentication part-specific color information 333, for example, the identification information for identifying a person and the authentication color information which is a registered feature amount are associated with each other. Here, the identification information included in the color information 333 for each part for authentication is the same as the identification information included in the face feature amount information 331. In addition, the color information for authentication includes color information for each part. For example, the first line of FIG. 3 includes color information (RGB (red, green, blue) information) of the right shoulder, left elbow, ....

The authentication color information in the authentication part-specific color information 333 may include color information of all parts that can be recognized by the skeleton recognition unit 343, which will be described later, or color information of some parts. Only may be included. The authentication color information preferably includes color information of a plurality of parts, but may include only color information of one part such as only the right elbow. For example, the color information for authentication includes color information of a characteristic part of a person's clothing, and color information of a part that must be worn such as long sleeves and a tie and must be compared by the clothing authentication unit 345 (for example, long sleeves). In the case of, the color information of the left and right wrists. In the case of a tie, the color information of the upper part of the spine), etc. can be included. In the authentication color information in the authentication site-specific color information 333, the color may be defined by other than RGB.

The image information 334 includes image data acquired by the camera 200. In the image information 334, for example, the image data and the information indicating the date and time when the camera 200 acquired the image data are associated with each other.

The skeleton coordinate information 335 includes information indicating the coordinates of each part of the person recognized by the skeleton recognition unit 343. FIG. 4 shows an example of the skeleton coordinate information 335. With reference to FIG. 4, in the skeleton coordinate information 335, the identification information and the part coordinates are associated with each other. The identification information is the same as the identification information included in the face feature amount information 331 and the authentication site-specific color information 333. In addition, the part coordinates include information indicating the coordinates of each part in the image data.

The part included in the part coordinates corresponds to the trained model 332. For example, in FIG. 4, the upper part of the spine, the right shoulder, the left shoulder, ..., Are illustrated. The site coordinates can include, for example, about 30 sites (other than those illustrated). The part included in the part coordinates may be other than those illustrated in FIG. 4 and the like.

The color information 336 for each part is the color information of each part acquired from the image data by the color information acquisition unit 344 for each part based on the image data included in the image information 334 and the part coordinates included in the skeleton coordinate information 335. Contains information that indicates. FIG. 5 shows an example of the color information 336 for each part. Referring to FIG. 5, in the part-specific color information 336, the identification information and the part color information are associated with each other. The identification information is the same as the identification information included in the face feature amount information 331, the color information 333 for each authentication part, and the skeleton coordinate information 335. Further, the part color information includes the color information of each part in the image data. For example, the first line of FIG. 5 includes color information (RGB (red, green, blue) information) of the upper part of the spine, the right shoulder, ....

The part color information in the part-specific color information 336 may include the color information of all the parts included in the skeleton coordinate information 335, or includes only the color information of some parts. It doesn't matter. For example, the part color information may include the color information of the part included in the authentication color information corresponding to the identification information. Further, in the part color information in the part-specific color information 336, the color may be defined by other than RGB as in the case of the authentication color information. However, it is desirable that the color of the part color information and the color information for authentication are defined by using the same method.

The authentication result information 337 includes information indicating the result of the authentication process by the clothing authentication unit 345. For example, in the authentication result information 337, the identification information and the information indicating whether or not to allow admission are associated with each other. The details of the authentication process by the clothing authentication unit 345 will be described later.

The arithmetic processing unit 340 has a microprocessor such as an MPU and its peripheral circuits. The arithmetic processing unit 340 reads the program 338 from the storage unit 330 and executes it, thereby realizing various processing units in cooperation with the hardware and the program 338. The main processing units realized by the arithmetic processing unit 340 include, for example, an image acquisition unit 341, a face authentication unit 342, a skeleton recognition unit 343, a part-specific color information acquisition unit 344, a clothing authentication unit 345, and an admission permission unit 346. There is.

The image acquisition unit 341 acquires the image data acquired by the camera 200 from the camera 200 via the communication I / F unit 320. Then, the image acquisition unit 341 stores the acquired image data in the storage unit 330 as image information 334 in association with, for example, the acquisition date and time of the image data.

The face authentication unit 342 performs face authentication based on the image data included in the image information 334. For example, the face authentication unit 342 extracts a person to be authenticated from the image data. In addition, the face recognition unit 342 extracts feature points such as eyes, nose, and mouth of the extracted person, and calculates a feature amount based on the extracted result. Then, the face recognition unit 342 calculates the feature amount by examining whether or not the similarity between the calculated feature amount and the face feature amount included in the face feature amount information 331 exceeds the face comparison threshold. Is collated with the feature amount stored in the storage unit 330, and authentication is performed based on the collation result. For example, the face recognition unit 342 identifies the identification information of the person to be authenticated by performing face recognition.

The skeleton recognition unit 343 recognizes the skeleton of the person to be authenticated by the face recognition unit 342 in the image data using the trained model 332. For example, as shown in FIG. 6, the skeletal recognition unit 343 recognizes each part such as the upper part of the spine, the right shoulder, the left shoulder, the right elbow, the left elbow, the right wrist, the left wrist, the right hand, the left hand, and so on. In addition, the skeleton recognition unit 343 calculates the coordinates in the screen data of each recognized part. Then, the skeleton recognition unit 343 associates the recognition / calculation result with the identification information specified by the authentication by the face recognition unit 342 and stores it in the storage unit 330 as the skeleton coordinate information 335.

The part recognized by the skeleton recognition unit 343 corresponds to the trained model 332 (teacher data used when learning the trained model 332). Therefore, the skeleton recognition unit 343 may recognize a part other than the above-exemplified parts according to the trained model 332.

The part-specific color information acquisition unit 344 acquires color information from the image data authenticated by the face authentication unit 342 by using the skeleton coordinate information 335 which is the result of recognition by the skeleton recognition unit 343. For example, as shown in FIG. 6, the part-specific color information acquisition unit 344 acquires color information at the coordinates of each part included in the skeleton coordinate information 335 among the image data. Then, the part-specific color information acquisition unit 344 stores the acquired color information and the identification information specified by the authentication by the face recognition unit 342 in the storage unit 330 as the part-specific color information 336 in association with each other.

Note that the color information acquisition unit 344 for each part may acquire the color information of all the coordinates included in the skeleton coordinate information 335, or may acquire the color information of some coordinates. For example, the part-specific color information acquisition unit 344 acquires only the color information at the coordinates of the part included in the authentication color information associated with the identification information specified by the face authentication unit 342 in the authentication part-specific color information 333. It may be configured to do so. Further, the color information acquisition unit 344 for each part may acquire the color information of the coordinates in the image data pinpointly, or may be configured to acquire the color information in a predetermined range centered on the coordinates, for example. .. When acquiring the color information in a predetermined range, the part-specific color information acquisition unit 344 uses, for example, the color information specified according to the average of the acquired color information or the ratio of the acquired color information as the color information of the part. You can estimate it.

The clothing authentication unit 345 compares the color information of each part acquired by the part-specific color information acquisition unit 344 with the color information of each part included in the authentication part-specific color information 333. In other words, as shown in FIG. 7, the clothing authentication unit 345 includes the part color information in the part-specific color information 336 and the authentication color information in the authentication part-specific color information 333 in which the identification information is the same. Is compared for each part. For example, in the case shown in FIG. 7, the clothing authentication unit 345 compares each part of the right shoulder, the left elbow, the left hand, and so on. Then, the clothing certification unit 345 performs authentication to determine whether or not the clothing is registered in advance based on the result of the comparison. After that, the clothing authentication unit 345 stores the information indicating the authentication result in the storage unit 330 as the authentication result information 337.

For example, the clothing certification unit 345 compares the part color information with the authentication color information by calculating the color distance indicated by the color information for each part. That is, the clothing certification unit 345 compares the part color information with the authentication color information by observing the color difference for each part. When the color distance is within a predetermined threshold value, the clothing authentication unit 345 determines that the color information of the part acquired by the part-specific color information acquisition unit 344 is registered in advance for the part. On the other hand, when the color distance exceeds a predetermined threshold value, the clothing authentication unit 345 determines that the color information of the part acquired by the part-specific color information acquisition unit 344 is not registered in advance for the part.

In addition, the clothing certification unit 345 performs authentication to determine whether or not the entire clothing is registered in advance based on the result of comparison of each part. For example, the clothing certification unit 345 has pre-registered the color information of the parts acquired by the part-specific color information acquisition unit 344 for all the parts included in the authentication color information in the authentication part-specific color information 333. If it is determined that there is, it is determined and authenticated that the clothing is registered in advance.

Note that the certification by the clothing certification unit 345 may be other than those illustrated above. For example, when the clothing certification unit 345 determines that among the parts included in the color information for authentication in the color information 333 for each part for authentication, more than a predetermined number of parts are registered in advance. It is possible to identify and authenticate that the clothes are registered in advance. Further, for example, the clothing certification unit 345 has determined that among the parts included in the color information for authentication in the color information 333 for each part for authentication, the parts defined in advance as essential are registered in advance. In this case, it can be determined and authenticated that the clothes are registered in advance. The authentication by the clothing authentication unit 345 may be discriminated and authenticated by, for example, a score calculated based on a value defined in advance for each part. For example, as described above, various conditions can be adopted when the clothing authentication unit 345 makes a determination / authentication.

The admission permission unit 346 performs processing based on the result of authentication by the clothing certification unit 345. For example, the admission permission unit 346 determines that the clothing is pre-registered by the clothing authentication unit 345, opens the gate, door, etc., and determines that the clothing is pre-registered in the screen display unit 310. It is possible to perform admission processing such as displaying the fact that the clothes have been authenticated, and transmitting the fact that the clothes have been determined and authenticated in advance to an external device via the communication I / F unit 320. Further, the admission permission unit 346 is not authenticated by transmitting an alarm to an external device via the communication I / F unit 320 when the clothing authentication unit 345 determines and authenticates that the clothing is not registered in advance. It may be configured to perform the processing in the case of the above.

The above is a configuration example of the authentication device 300.

Subsequently, the operation of the authentication device 300 will be described with reference to FIG. FIG. 8 is a flowchart showing an operation example of the authentication device 300. Referring to FIG. 8, the face recognition unit 342 of the authentication device 300 performs face recognition based on the image data included in the image information 334 (step S101). The face recognition process performed by the face recognition unit 342 may be a general one. Then, the face recognition unit 342 identifies the identification information of the person to be authenticated as a result of the face recognition.

The skeleton recognition unit 343 recognizes the skeleton of the person to be authenticated by the face recognition unit 342 in the image data using the trained model 332. In addition, the skeleton recognition unit 343 calculates the coordinates in the screen data of each recognized part. For example, as described above, the skeleton recognition unit 343 specifies the coordinates of each part by performing skeleton recognition (step S102).

The part-specific color information acquisition unit 344 uses the skeleton coordinate information 335, which is the result of recognition by the skeleton recognition unit 343, to obtain the color of each part indicated by the skeleton coordinate information 335 from the image data authenticated by the face recognition unit 342. Acquire information (step S103).

Note that the color information acquisition unit 344 for each part may acquire the color information of all the coordinates included in the skeleton coordinate information 335, or may acquire the color information of some coordinates. For example, the part-specific color information acquisition unit 344 acquires only the color information at the coordinates of the part included in the authentication color information associated with the identification information specified by the face recognition unit 342 in the authentication part-specific color information 333. It doesn't matter.

The clothing authentication unit 345 compares the color information of each part acquired by the part-specific color information acquisition unit 344 with the color information of each part included in the authentication part-specific color information 333 (step S104). For example, the clothing certification unit 345 compares the part color information with the authentication color information by calculating the distance of the color indicated by the color information for each part.

Further, the clothing authentication unit 345 determines and authenticates whether or not the clothing of the person is registered in advance based on whether or not the result of comparison for each part satisfies a predetermined condition (step S105). ). For example, when the result of comparison for each part satisfies a predetermined condition (step S105, Yes), the clothing authentication unit 345 determines that the clothing of the person to be authenticated is registered in advance. In this case, the admission permission unit 346 performs admission processing such as admitting admission (step S106). On the other hand, if the result of comparison for each part does not satisfy the predetermined conditions (step S105, No), the admission permission unit 346 does not allow admission.

The above is an operation example of the authentication device 300.

As described above, the authentication device 300 has a skeleton recognition unit 343, a site-specific color information acquisition unit 344, and a clothing authentication unit 345. With such a configuration, the part-specific color information acquisition unit 344 can acquire the color information of the part by using the skeleton coordinate information 335 which is the result of the recognition by the skeleton recognition unit 343. In addition, the clothing authentication unit 345 can perform authentication based on the color information of the part acquired by the part-specific color information acquisition unit 344. As a result, it becomes possible to compare the color information of an accurate area for performing clothing authentication. This makes it possible to improve the accuracy of clothing authentication.

Further, according to the above configuration, it is possible to perform authentication based on the color information of a plurality of parts acquired by the part-specific color information acquisition unit 344. By performing authentication based on the color information of a plurality of parts in this way, it is possible to perform more detailed clothing authentication such as whether the person is wearing a designated tie or has long sleeves. As a result, it is possible to improve the accuracy of clothing authentication.

In the present embodiment, an example is given in the case where an individual is identified based on the result of face authentication. That is, the case where the target specific information for specifying the color information 333 for each part for authentication, which is the registered feature amount, is acquired by face authentication is illustrated. However, for example, as shown in FIG. 9, an RFID (radio frequency identifier) tag such as an employee ID card may be used to identify an individual. In the case shown in FIG. 9, the authentication device 300 has an ID information acquisition unit 347 instead of the face authentication unit 342, and the storage unit 330 stores ID information 339 instead of the face feature amount information 331. .. The ID information acquisition unit 347 receives ID information from an external RFID reader or the like via the communication I / F unit 320 or the like. Further, the ID information acquisition unit 347 can specify the identification information based on the received ID information and the ID information 339 that associates the ID information with the identification information. For example, as shown in FIG. 9, the authentication device 300 may be configured to acquire ID information instead of face authentication. The authentication device 300 may specify the identification information by a known authentication process such as iris authentication or fingerprint authentication instead of the face authentication process.

Further, the authentication device 300 may be configured so as not to identify an individual. For example, the authentication device 300 is configured to discriminate attributes such as age and gender based on image data, and perform clothing authentication using authentication color information associated with a group according to the discriminated attributes. It doesn't matter. Further, the clothing authentication unit 345 may be configured to perform clothing authentication using predetermined authentication color information without performing face authentication or attribute determination. In this way, the authentication device 300 does not necessarily have to identify and discriminate a group according to an individual or an attribute. In addition, in the case of performing clothing authentication for each group according to the attribute as described above, in the color identification information 333 for each authentication part, the color information for authentication is associated with each predetermined group in which a plurality of identification information are collected. It doesn't matter if it is done. The grouping can be set in advance according to the attributes of the person such as job title, occupation, gender, and the like.

Further, in the present embodiment, a case where the admission permission unit 346 performs admission processing in the case of pre-registered clothes has been described. However, the result of the certification by the clothing certification unit 345 may be used for purposes other than admission processing. For example, the authentication device 300 may be configured to refuse admission when it is determined and authenticated that the clothing is pre-registered as a result of authentication by the clothing authentication unit 345. As described above, the processing performed as a result of the authentication by the clothing authentication unit 345 may be other than the admission permission by the admission permission unit 346.

Further, in the present embodiment, a case where skeleton recognition is performed based on image data is illustrated. However, skeleton recognition may be performed using, for example, an infrared depth sensor.

Further, in the present embodiment, a case where color information is used as a feature amount used when performing clothing authentication has been described. However, the feature amount may be other than the color information. The feature amount can include, for example, a value according to the pattern or pattern of the clothes.

Further, in FIGS. 1 and 2, a case where the function as the authentication device 300 is realized by one information processing device has been described. However, the function as the authentication device 300 may be realized by, for example, a plurality of information processing devices connected via a network. Further, FIG. 1 illustrates a case where the authentication system 100 is composed of a camera 200 and an authentication device 300. However, the camera 200 and the authentication device 300 may be integrally configured, such as the camera 200 having the authentication device 300 and the authentication device 300 having the function as the camera 200. For example, as described above, the configuration of the authentication system 100 is not limited to the case illustrated in FIG. 1, and various modified examples can be adopted.

[Second Embodiment]
Next, a second embodiment of the present invention will be described with reference to FIGS. 10 and 11. 10 and 11 show a configuration example of the authentication device 400.

FIG. 10 shows a hardware configuration example of the authentication device 400. Referring to FIG. 10, the authentication device 400 has the following hardware configuration as an example.
-CPU (Central Processing Unit) 401 (arithmetic unit)
-ROM (Read Only Memory) 402 (storage device)
-RAM (Random Access Memory) 403 (storage device)
-Program group 404 loaded into RAM 403
A storage device 405 that stores a program group 404.
-Drive device 406 that reads and writes the recording medium 410 outside the information processing device
-Communication interface 407 that connects to the communication network 411 outside the information processing device
-I / O interface 408 that inputs and outputs data
-Bus 409 connecting each component

Further, the authentication device 400 can realize the functions as the acquisition unit 421 and the authentication unit 422 shown in FIG. 11 by the CPU 401 acquiring the program group 404 and executing the program group 404. The program group 404 is stored in, for example, a storage device 405 or a ROM 402 in advance, and the CPU 401 loads the program group 404 into a RAM 403 or the like and executes the program group 404 as needed. Further, the program group 404 may be supplied to the CPU 401 via the communication network 411, or may be stored in the recording medium 410 in advance, and the drive device 406 may read the program and supply the program to the CPU 401.

Note that FIG. 10 shows an example of a hardware configuration of the authentication device 400. The hardware configuration of the authentication device 400 is not limited to the above case. For example, the authentication device 400 may be configured from a part of the above-described configuration, such as not having the drive device 406.

The acquisition unit 421 acquires the feature amount of the part of the person specified by recognizing the skeleton of the person in the image data by skeleton recognition from the image data.

Based on the feature amount acquired by the acquisition unit 421, the authentication unit 422 performs clothing authentication for determining whether or not the person in the image data is the clothing registered in advance.

As described above, the authentication device 400 has an acquisition unit 421 and an authentication unit 422. With such a configuration, the authentication unit 422 can perform clothing authentication for determining whether or not the person in the image data is the clothing registered in advance based on the feature amount acquired by the acquisition unit 421. As a result, it becomes possible to compare the color information of an accurate area for performing clothing authentication. This makes it possible to improve the accuracy of clothing authentication.

The above-mentioned authentication device 400 can be realized by incorporating a predetermined program into the authentication device 400. Specifically, the program, which is another embodiment of the present invention, causes the authentication device 400 to obtain the feature amount of the part of the person specified by recognizing the skeleton of the person in the image data by skeleton recognition from the image data. The acquisition unit 421 to be acquired and the authentication unit 422 that performs clothing authentication for determining whether or not the person in the image data is the clothing registered in advance based on the feature amount acquired by the acquisition unit 421 are realized. It is a program for.

Further, in the authentication method executed by the above-mentioned authentication device 400, the authentication device 400 determines the feature amount of the part of the person specified by recognizing the skeleton of the person in the image data by skeleton recognition from the image data. It is a method of acquiring and performing clothing authentication for determining whether or not a person in the image data is a pre-registered clothing based on the acquired feature amount.

Even an invention of a program or an authentication method having the above-mentioned configuration can achieve the above-mentioned object of the present invention because it has the same action and effect as the above-mentioned authentication device 400.

<Additional notes>
Part or all of the above embodiments may also be described as in the appendix below. Hereinafter, the outline of the authentication method and the like in the present invention will be described. However, the present invention is not limited to the following configurations.

(Appendix 1)
From the image data, an acquisition unit that acquires the feature amount of the part of the person specified by recognizing the skeleton of the person in the image data by skeleton recognition, and
Based on the acquired feature amount, an authentication unit that performs clothing authentication to determine whether or not the person in the image data is the pre-registered clothing, and an authentication unit.
Authentication device with.
(Appendix 2)
The authentication device described in Appendix 1
The acquisition unit identifies a plurality of parts of the person by recognizing the skeleton of the person in the image data by skeleton recognition, and acquires the feature amounts of the specified plurality of parts.
The authentication unit is an authentication device that authenticates clothes by comparing the acquired feature amount with the registered feature amount registered in advance for each part for each part.
(Appendix 3)
The authentication device described in Appendix 2
The authentication unit is an authentication device that makes a comparison for each part by calculating the distance between the feature amount and the registered feature amount.
(Appendix 4)
The authentication device according to Appendix 2 or Appendix 3.
An authentication device in which the feature amount and the registered feature amount are color information.
(Appendix 5)
The authentication device according to any one of Supplementary note 2 to Supplementary note 4.
Acquire the target identification information for specifying the registered feature amount,
The authentication unit is an authentication device that authenticates clothes by comparing the acquired feature amount with the registered feature amount specified based on the target identification information.
(Appendix 6)
The authentication device according to Appendix 5.
The registered feature amount is registered for each of the groups when the person is divided into a plurality of groups according to the attributes of the person.
The target identification information is an authentication device that is information for identifying the group.
(Appendix 7)
The authentication device according to Appendix 5.
The registered feature amount is registered for each person,
Target identification information is an authentication device that is information for identifying a person.
(Appendix 8)
The authentication device according to any one of Supplementary note 5 to Supplementary note 7.
An authentication device that acquires the target identification information by performing face recognition based on image data.
(Appendix 9)
Information processing device
From the image data, the feature amount of the part of the person specified by recognizing the skeleton of the person in the image data by skeleton recognition is acquired.
An authentication method for performing clothing authentication for determining whether or not a person in image data is dressed in advance based on the acquired feature amount.
(Appendix 10)
For information treatment equipment,
From the image data, an acquisition unit that acquires the feature amount of the part of the person specified by recognizing the skeleton of the person in the image data by skeleton recognition, and
Based on the feature amount acquired by the acquisition unit, an authentication unit that performs clothing authentication for determining whether or not a person in the image data is a pre-registered clothing, and an authentication unit.
A computer-readable recording medium that records programs to achieve this.

Note that the programs described in each of the above embodiments and appendices may be stored in a storage device or recorded in a recording medium readable by a computer. For example, the recording medium is a portable medium such as a flexible disk, an optical disk, a magneto-optical disk, and a semiconductor memory.

Although the invention of the present application has been described above with reference to each of the above embodiments, the invention of the present application is not limited to the above-described embodiment. Various changes that can be understood by those skilled in the art can be made to the structure and details of the present invention within the scope of the present invention.

The present invention enjoys the benefit of priority claim based on the patent application of Japanese Patent Application No. 2020-037821 filed in Japan on March 5, 2020, and is described in the patent application. All contents are included in this specification.

100 Authentication system 200 Camera 300 Authentication device 310 Screen display unit 320 Communication I / F unit 330 Storage unit 331 Face feature amount information 332 Learned model 333 Authentication site-specific color information 334 Image information 335 Skeletal coordinate information 336 Site-specific color information 337 Authentication result information 338 Program 339 ID information 340 Calculation processing unit 341 Image acquisition unit 342 Face recognition unit 343 Skeleton recognition unit 344 Part-specific color information acquisition unit 345 Clothing authentication unit 346 Admission permission unit 347 ID information acquisition unit 401 CPU
402 ROM
403 RAM
404 Program group 405 Storage device 406 Drive device 407 Communication interface 408 Input / output interface 409 Bus 410 Recording medium 411 Communication network 421 Acquisition unit 422 Authentication unit

Claims

From the image data, an acquisition unit that acquires the feature amount of the part of the person specified by recognizing the skeleton of the person in the image data by skeleton recognition, and
Based on the acquired feature amount, an authentication unit that performs clothing authentication to determine whether or not the person in the image data is the pre-registered clothing, and an authentication unit.
Authentication device with.
The authentication device according to claim 1.
The acquisition unit identifies a plurality of parts of the person by recognizing the skeleton of the person in the image data by skeleton recognition, and acquires the feature amounts of the specified plurality of parts.
The authentication unit is an authentication device that authenticates clothes by comparing the acquired feature amount with the registered feature amount registered in advance for each part for each part.
The authentication device according to claim 2.
The authentication unit is an authentication device that makes a comparison for each part by calculating the distance between the feature amount and the registered feature amount.
The authentication device according to claim 2 or 3.
An authentication device in which the feature amount and the registered feature amount are color information.
The authentication device according to any one of claims 2 to 4.
Acquire the target identification information for specifying the registered feature amount,
The authentication unit is an authentication device that authenticates clothes by comparing the acquired feature amount with the registered feature amount specified based on the target identification information.
The authentication device according to claim 5.
The registered feature amount is registered for each of the groups when the person is divided into a plurality of groups according to the attributes of the person.
The target identification information is an authentication device that is information for identifying the group.
The authentication device according to claim 5.
The registered feature amount is registered for each person,
Target identification information is an authentication device that is information for identifying a person.
The authentication device according to any one of claims 5 to 7.
An authentication device that acquires the target identification information by performing face recognition based on image data.
Information processing device
From the image data, the feature amount of the part of the person specified by recognizing the skeleton of the person in the image data by skeleton recognition is acquired.
An authentication method for performing clothing authentication for determining whether or not a person in image data is dressed in advance based on the acquired feature amount.
For information treatment equipment,
From the image data, an acquisition unit that acquires the feature amount of the part of the person specified by recognizing the skeleton of the person in the image data by skeleton recognition, and
Based on the feature amount acquired by the acquisition unit, an authentication unit that performs clothing authentication for determining whether or not a person in the image data is a pre-registered clothing, and an authentication unit.
A computer-readable recording medium that records programs to achieve this.