Image processing device, image processing method, recording medium, and authentication system

Abstract

This image processing device comprises: a determination unit that identifies the quality of a face image and determines the necessity of correction; a display processing unit that displays, in association with each other, a first face image for which correction is determined to be necessary and correction information with which recommended correction processing can be identified; and a registration unit that, when the determination input for performing the correction processing is performed, registers a second face image obtained after the correction processing has been performed on the first face image.

Description

Image processing device, image processing method, recording medium, and authentication system

[0001] The present invention relates to an image processing device, an image processing method, a program, and an authentication system.

[0002] Patent Literature 1 describes a technology for improving robustness against individual differences in facial features or fluctuations in facial orientation in a facial authentication system by identifying an individual using additional facial images of the individual wearing an accessory. Specifically, Patent Literature 1 generates facial images of an individual corresponding to a plurality of different facial orientations, and further generates facial images of the individual wearing an accessory such as a mask for each of the generated facial images, and identifies the individual using these plurality of types of facial images.

[0003] JP 2023-6150 A

[0004] In the system described in Patent Literature 1, multiple types of facial images of individuals are prepared to improve the accuracy of personal identification, but the system does not disclose, for example, a procedure in which an operator checks and registers facial images before and after correction. In an authentication system, if a registered facial image is not suitable for authentication processing, authentication accuracy decreases. Therefore, in the authentication system, it is necessary to check whether a registered facial image is suitable for authentication processing and correct it as appropriate. However, the greater the number of facial images to be authenticated, the more labor-intensive these tasks become.

[0005] The image processing device of the present disclosure includes a determination means for identifying the quality of a facial image and determining whether or not correction is required; a display processing means for displaying a first facial image determined to require correction in association with correction information that can identify a recommended correction process; and a registration means for registering a second facial image after the correction process has been performed on the first facial image when a decision to perform the correction process is input.

[0006] The authentication system in the present disclosure includes an authentication device that performs authentication processing using a facial image, and an image processing device that processes the facial image used by the authentication device for the authentication processing, wherein the image processing device has: a determination means that identifies the quality of the facial image and determines whether or not correction is necessary; a display processing means that displays a first facial image that is determined to require correction in association with correction information that can identify a recommended correction process; and a registration means that, when a decision input is made to perform the correction process, registers a second facial image after the correction process has been performed on the first facial image.

[0007] The image processing method of the present disclosure includes one or more computers: identifying the quality of a facial image, determining whether or not correction is necessary; displaying a first facial image determined to require correction in association with correction information that can identify a recommended correction process; and, when a decision to perform the correction process is input, registering a second facial image after the correction process has been performed on the first facial image.

[0008] The program in the present disclosure causes a computer to execute a determination process of identifying the quality of a facial image and determining whether or not correction is necessary; a display process of displaying a first facial image determined to require correction in association with correction information that can identify a recommended correction process; and a registration process of registering a second facial image after the correction process has been performed on the first facial image when a decision to perform the correction process is input.

[0009] According to the present disclosure, an image processing device, an image processing method, a program, and an authentication system are provided that can reduce the effort required to modify a face image used in authentication processing to one suitable for authentication processing.

[0010] FIG. 1 is a functional block diagram showing an example configuration of an image processing device according to the present disclosure. FIG. 2 is a flowchart showing an example processing operation of an image processing device according to the present disclosure. FIG. 3 is a diagram conceptually showing an example system configuration of an authentication system according to the present disclosure. FIG. 4 is a diagram showing an example configuration of a computer that realizes an image processing device according to the present disclosure. FIG. 5 is a diagram showing an example data structure of image quality information and image information. FIG. 6 is a diagram showing an example of a registration face list screen. FIG. 7 is a diagram showing an example of the registration face list screen after image correction. FIG. 7 is a flowchart showing an example processing operation of an image processing device according to the present disclosure. FIG. 8 is a flowchart showing an example processing operation of an image processing device according to the present disclosure. FIG. 9 is a diagram showing an example correction screen. FIG. 10 is a diagram showing an example correction screen where detailed settings of correction items can be made. FIG. 11 is a diagram showing an example of a simple correction window. FIG. 11 is a diagram showing an example of a registration face batch correction screen. FIG. 12 is a diagram showing an example of a registration face confirmation screen.

[0011] Hereinafter, in this disclosure, the drawings relate to one or more embodiments. In all drawings, similar components are denoted by similar reference numerals, and descriptions thereof will be omitted as appropriate. In each of the following drawings, configurations of parts that are not related to the essence of this disclosure are omitted and are not shown.

[0012] In the present disclosure, "acquisition" includes at least one of the following: a device going to retrieve data or information stored in another device or storage medium (active acquisition); and a device inputting data or information output from another device (passive acquisition). Examples of active acquisition include making a request or inquiry to another device and receiving a response, and accessing and reading information from another device or storage medium. Examples of passive acquisition include receiving information that is distributed (or transmitted, push notification, etc.). Furthermore, "acquisition" may also mean selecting and acquiring data or information from received data or information, or selecting and receiving distributed data or information.

[0013] <Example of Functional Configuration> As shown in Fig. 1, the image processing device 100 includes a determination unit 102, a display processing unit 104, and a registration unit 106. The determination unit 102 identifies the quality of a facial image and determines whether or not retouching is necessary. The display processing unit 104 displays a first facial image determined to require retouching in association with retouching information that can identify a recommended retouching process. When a decision to perform retouching is input, the registration unit 106 registers a second facial image obtained after the retouching process has been performed on the first facial image.

[0014] <Operation Example> The image processing device 100 operates as shown in the flowchart of Fig. 2. First, the determination unit 102 identifies the quality of the facial image (step S101) and determines whether retouching is necessary (step S103). The display processing unit 104 displays the first facial image determined to require retouching in association with retouching information that can identify the recommended retouching process (step S105). When a decision to perform retouching is input (YES in step S107), the registration unit 106 registers the second facial image after the retouching process has been performed on the first facial image (step S109) (step S111).

[0015] According to this image processing device 100, the determination unit 102 identifies the quality of the facial image and determines whether or not retouching is necessary. The display processing unit 104 displays a first facial image determined to require retouching in association with retouching information that can identify the recommended retouching process. When a decision to perform retouching is input, the registration unit 106 registers a second facial image obtained after the retouching process has been performed on the first facial image. This reduces the effort required to retouch a facial image used in authentication processing so that it is suitable for authentication processing.

[0016] A detailed example of the image processing device 100 will be described below.

[0017] (First embodiment) <System overview> Fig. 3 is a diagram conceptually illustrating the system configuration of an authentication system 1 according to the present disclosure. The authentication system 1 includes an image processing device 100 and an authentication device 200. The image processing device 100 and the authentication device 200 may or may not be connected to each other via a communication network. The authentication device 200 performs authentication processing using a facial image. However, the authentication device 200 may also perform authentication processing using an image other than a facial image. For example, the authentication device 200 may also perform authentication processing using registered authentication information including at least one feature such as iris, veins, auricle, fingerprint, voiceprint, gait, and physique (height, shoulder width, body length, bone structure, etc.).

[0018] The registered authentication information is associated with information (e.g., identification information) capable of identifying the authentication target and is registered in the storage device 120. The authentication device 200 performs authentication processing of the authentication target by comparing features extracted from the registered authentication information including a facial image stored in the storage device 120 with features extracted from the authentication target and corresponding to the registered authentication information. However, the storage device 120 may store features extracted from a facial image as registered authentication information.

[0019] The feature data can be extracted using, but not limited to, an algorithm such as SIFT (Scale-Invariant Feature Transform) or HOG (Histograms of Oriented Gradients).

[0020] Because authentication processing is used in various situations, the details of the authentication processing are not particularly limited. The following authentication processing details are an example. For example, when authentication processing is performed at the time of gate entry, authentication device 200 acquires identification information recorded on the ID card of the person to be authenticated and also acquires a facial image generated by photographing the face of the person to be authenticated using a camera or the like. Then, authentication device 200 compares features extracted from the acquired facial image of the person to be authenticated with features extracted from registered authentication information associated with the acquired identification information of the person to be authenticated. If the comparison result satisfies a criterion, for example, if the degree of match between the two features is equal to or greater than a threshold (hereinafter, "above the threshold" may also be "above the threshold"), authentication device 200 determines that the person to be authenticated is the person to be authenticated. As a result, authentication of the person to be authenticated is successful, and the gate is opened. On the other hand, if the comparison result does not satisfy the criterion, for example, if the degree of match between the two features is less than a threshold (if the criterion is "above the threshold," it is "below the threshold"), authentication device 200 determines that the person to be authenticated is not the person to be authenticated. As a result, the authentication of the authentication subject fails and the gate is closed.

[0021] Furthermore, the face image corrected by the image processing device 100 may also be used to track a specific person by the authentication system 1. The authentication device 200 of the authentication system 1 may track a specific person by using a registered face image of the specific person as the person to be authenticated and performing a matching process with a face image of a person captured in a video image from a surveillance camera or the like.

[0022] The image processing device 100 performs processing to improve the accuracy of the facial image of the person to be authenticated that is used in the authentication process by the authentication device 200. The facial image of the person to be authenticated that has been processed by the image processing device 100 is registered in the storage device 120 and is used in the authentication process by the authentication device 200.

[0023] 3, the image processing device 100 and the authentication device 200 are shown as separate pieces of hardware, but the image processing device 100 may be provided either inside or outside the authentication device 200. The image processing device 100 may realize the functions of the image processing device 100 on the authentication device 200 by installing an application as an image processing tool in the authentication device 200.

[0024] The image processing device 100 is a computer such as a smartphone, a tablet terminal, a personal computer, a server computer, etc. The authentication device 200 is a computer such as a personal computer, a server computer, etc. However, the authentication device 200 may be implemented in a form incorporated into a device that performs authentication processing, such as a gate device or a surveillance camera.

[0025] The storage device 120 may be provided inside or outside the image processing device 100 or the authentication device 200. In other words, the storage device 120 may be hardware that is integrated with the image processing device 100 or the authentication device 200, or may be hardware that is separate from the image processing device 100 or the authentication device 200.

[0026] Furthermore, the storage device 120 may be composed of multiple storage devices. The storage device 120 may store, in addition to a facial image used in authentication processing (corresponding to the second facial image described above), images before image processing by the image processing device 100 (images including an image corresponding to the first facial image described above). The second facial image and the first facial image may be stored in two different storage devices, respectively, or may be stored in different storage areas of a single storage device. For example, in the former case, the storage device that stores the first facial image may be connected in a state accessible to the image processing device 100, and the storage device that stores the second facial image may be connected in a state accessible to the image processing device 100 and the authentication device 200.

[0027] <Hardware Configuration Example> Fig. 4 is a block diagram illustrating an example of the hardware configuration of a computer 1000 that realizes the image processing device 100 according to the present disclosure. The authentication device 200 in Fig. 3 is also realized by the computer 1000. As described above, the image processing device 100 and the authentication device 200 may be realized by the same computer 1000.

[0028] The computer 1000 has a bus 1010 , a processor 1020 , a memory 1030 , a storage device 1040 , an input / output interface 1050 , and a network interface 1060 .

[0029] The bus 1010 is a data transmission path for transmitting and receiving data among the processor 1020, memory 1030, storage device 1040, input / output interface 1050, and network interface 1060. However, the method of connecting the processor 1020 and the like to each other is not limited to bus connection.

[0030] The processor 1020 is implemented by a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), or the like.

[0031] The memory 1030 is a main storage device realized by a RAM (Random Access Memory) or the like.

[0032] The storage device 1040 is an auxiliary storage device realized by a hard disk drive (HDD), a solid state drive (SSD), a memory card, a read only memory (ROM), or the like. The storage device 1040 stores program modules that realize each function of the image processing device 100 (e.g., the determination unit 102, the display processing unit 104, the registration unit 106, etc.). The processor 1020 loads each of these program modules into the memory 1030 and executes them, thereby realizing each function corresponding to the program module. The storage device 1040 may also function as the storage device 120 in FIG. 3 .

[0033] The program module may be recorded on a recording medium. The recording medium on which the program module is recorded may include a non-transitory, tangible medium usable by the computer 1000, and the program code readable by the computer 1000 (processor 1020) may be embedded in the medium.

[0034] The input / output interface 1050 is an interface for connecting the computer 1000 with various input / output devices. The input / output interface 1050 also functions as a communication interface for performing short-range wireless communication such as Bluetooth (registered trademark) and NFC (Near Field Communication).

[0035] The network interface 1060 is an interface for connecting the computer 1000 to a communication network. This communication network may be, for example, a local area network (LAN) or a wide area network (WAN). The network interface 1060 may connect to the communication network wirelessly or by wire.

[0036] The computer 1000 then connects to necessary devices (e.g., displays, touch panels, operation buttons, touchpads, keyboards, mice, speakers, microphones, cameras, printers, etc. of the image processing device 100 and authentication device 200) via the input / output interface 1050 or the network interface 1060.

[0037] Each component of the image processing device 100 according to the present disclosure in each figure is realized by any combination of hardware and software of the computer 1000 in Figure 4. Those skilled in the art will understand that there are various variations in the realization method and device. The functional block diagrams showing the image processing device 100 according to the present disclosure in each figure show blocks of logical functional units, rather than a configuration in hardware units.

[0038] <Example of Functional Configuration> The following will be described with reference to Fig. 1. The image processing device 100 includes a determination unit 102, a display processing unit 104, and a registration unit 106. The determination unit 102 identifies the quality of a facial image and determines whether or not correction is required. The facial image is an image that includes the face of a person to be authenticated.

[0039] <Method for Identifying Facial Image Quality> First, a method for identifying the quality of a facial image will be described below. Because a facial image is an original image used in the authentication process by authentication device 200, poor image quality may result in poor authentication accuracy. Therefore, it is preferable that the quality of the facial image be maintained at a certain standard so as not to reduce authentication accuracy.

[0040] The determination unit 102 determines the quality of a facial image in accordance with predetermined conditions. Various predetermined conditions are possible, including, but not limited to, the following examples. The quality of a facial image may be determined using at least one of the following conditions, or a combination of multiple conditions: (a1) The facial image contains a "face"; (a2) The detected "face" has a facial-likeness equal to or greater than a threshold. The facial-likeness can be determined, for example, based on the presence of each facial feature (eyes, eyebrows, nose, mouth, etc.) and their relative positions. (a3) The face is facing forward, or the deviation from the front is within a threshold. (a4) The face is not tilted, or the tilt is within a threshold. (a5) The brightness of the face meets the criteria (no shadows, not dark, not too bright and blown out, etc.). (a6) The face is not obscured (eyes not covered by hair, not wearing a mask or sunglasses, etc.). (a7) The facial expression meets the criteria (eyes open, mouth closed, etc.). (a8) The size and quality of the image meet the criteria (for example, the image size is above a threshold, the resolution is above a threshold, the subject is not blurred, etc.).

[0041] <Method for Determining Whether Modification is Necessary> The determination unit 102 identifies the quality of the facial image according to the above-described predetermined conditions and determines whether the quality of the facial image is suitable for authentication processing. For example, the determination unit 102 determines that modification is necessary if at least one of the above conditions is not satisfied. In other words, the determination unit 102 determines that modification is unnecessary if all of the above conditions are satisfied. However, it is not necessary to adopt all of the above conditions, and the conditions used to determine the quality may be selectable. Furthermore, the criteria (thresholds) for each condition may be changeable as appropriate. Furthermore, priorities (scores) may be set for the conditions used to determine the quality, and the determination unit 102 may determine that modification is unnecessary if the total value of the scores that satisfy the conditions is equal to or greater than a threshold.

[0042] Selection of conditions and setting changes of criteria and thresholds for each condition may be performed by accepting operator operations via a UI (User Interface) of a setting menu.

[0043] In the authentication system 1, the facial images registered as registered authentication information in the authentication device 200 are also selected using the same criteria as above before being registered. In the image processing device 100 of the present disclosure, it is preferable to use criteria that are stricter than the criteria used when registering facial images as registered authentication information in the authentication device 200, for example.

[0044] The determination unit 102 stores the determination result as image quality information 130 in the memory 1030, the storage device 1040, or the storage device 120 of the computer 1000, for example. As shown in FIG. 5 , the image quality information 130 includes, for example, identification information capable of identifying the person to be authenticated (shown as a user ID in the figure), identification information capable of identifying a facial image of the person to be authenticated (shown as an image ID in the figure), information indicating the result of quality determination, and a determination result, all associated with each other. The information indicating the result of quality determination is information indicating the result of quality determination based on the above-mentioned conditions. The information indicating the result of quality determination is, for example, whether a face was detected (e.g., "1" for face presence, "0" for no face), the proportion indicating a face-likeness, the proportion of faces facing forward, etc. The determination result is information indicating the determination result indicating whether correction is necessary based on the quality.

[0045] Furthermore, as shown in FIG. 5 , information on registered facial images for each person to be authenticated is stored as image information 140 in the memory 1030, storage device 1040, or storage device 120 of the computer 1000. For each person to be authenticated (user ID), the image information 140 includes identification information (image ID) of the facial image of the person to be authenticated, associated with information indicating at least one of the storage location (path) and file name of the facial image. Multiple facial images may be registered for one person to be authenticated. The facial image data file stored in the storage device 120 may be a data file in various formats, such as, but not limited to, JPEG format. Furthermore, the storage device 120 may store facial feature data extracted from the facial image in association with the facial image.

[0046] The display processing unit 104 displays the first facial image determined to require correction in association with correction information capable of identifying a recommended correction process. The correction information capable of identifying a recommended correction process includes, for example, at least one of information indicating that correction is recommended and information indicating whether correction is necessary. As an example, the correction information may be an image element such as an icon using a symbol, mark, or character capable of identifying whether correction is necessary. The correction information is displayed in a manner that makes the association with the first facial image clear. For example, the image element may be superimposed on the first facial image so that at least a portion of the image element overlaps with the first facial image, or may be displayed so that the image element is connected to the first facial image by a line, or may be displayed near the first facial image.

[0047] In another example, the correction information may be an image element that surrounds the first facial image using a different color or line type to distinguish whether or not correction is required. The correction information may also be an image element that indicates that correction is recommended, and the image element may be superimposed on the first facial image, with no image element being added to facial images other than the first facial image. For example, the display processing unit 104 may display the correction information in a way that emphasizes the first facial image itself (for example, by flashing it, applying a shading effect, or the like).

[0048] The display processing unit 104 displays at least two first face images on the same screen.

[0049] Specifically, the display processing unit 104 displays a registration face list screen 300 on a display (not shown) of the image processing device 100. As shown in FIG. 6 , the registration face list screen 300 has at least two face image display fields 310, and face images to be registered as authentication face images (or may be already registered face images) are displayed in the face image display fields 310. In the example of FIG. 6 , a no-editing face image 330 determined by the determination unit 102 as not requiring edits is displayed in the face image display field 310 together with a first indicator icon 320. The no-editing face image 330 may further be superimposed with a face area display 332 indicating a face area having a quality equal to or higher than a certain standard. However, as will be described later, the display processing unit 104 may also display the first face images one by one.

[0050] Furthermore, in the example of Fig. 6, the first facial image determined by the determination unit 102 to require correction is displayed as a correction-requiring facial image 340 in the facial image display field 310 together with the second indicator icon 322. As described above, the second indicator icon 322 is an example of correction information. The second indicator icon 322 can accept a press operation by the operator, and the processing performed by pressing the second indicator icon 322 will be described in a later-described embodiment. In this embodiment, the second indicator icon 322 realizes a function of informing the operator that correction is recommended for the facial image.

[0051] The first indicator icon 320 and the second indicator icon 322 are displayed superimposed on the bottom right corner of the face image display field 310, but this is just an example, and as mentioned above, the display position and display method are not limited to this. However, it is preferable that the second indicator icon 322 be in a form that stands out more than the first indicator icon 320.

[0052] In this way, on the registration face list screen 300, first face images that require correction (correction-requiring face images 340) are displayed together with the second indicator icon 322, and face images that do not require correction (correction-unnecessary face images 330) are displayed together with the first indicator icon 320. Therefore, the operator can distinguish at a glance between first face images that require correction and face images that do not require correction.

[0053] When a decision to perform the correction process is input, the registration unit 106 registers the second facial image obtained after the correction process has been performed on the first facial image in the storage device 120 as registered authentication information to be used by the authentication system 1. At this time, the registration unit 106 also updates, for example, the information in the image information 140 described above. The image information 140 may further include, in association with each facial image, information indicating whether the facial image is a facial image before correction or a facial image after correction.

[0054] The second facial image may be overwritten on the first facial image, but preferably, the unmodified first facial image is stored separately as an original image in a separate image file or in a separate storage area. Alternatively, an instruction as to whether to save or delete (overwrite with the second facial image) the first facial image may be received from an operator. The registration unit 106 may delete (overwrite with the second facial image) the first facial image in accordance with the received instruction, or may store the first facial image as a separate image file or in a separate storage area without deleting it. Alternatively, whether to save or delete (overwrite with the second facial image) the first facial image may be set in advance by receiving a selection from the operator via a setting menu. The registration unit 106 may delete (overwrite with the second facial image) the first facial image in accordance with the setting, or may store the first facial image as a separate file or in a separate storage area without deleting it.

[0055] Details of the correction process will be described later. After the correction process is performed, the display processing unit 104 displays, for example, a registration face list screen 300 as shown in FIG. 7 on the display. In this example, the registration face list screen 300 displays a correction-free face image 330 as the second face image after correction has been performed on the correction-required face image 340 (first face image) in FIG. 6. The second face image is displayed in the face image display field 310 in association with a third indicator icon 324. The third indicator icon 324 indicates that the second face image displayed in the face image display field 310 has been corrected or has become a face image of a certain quality as a result of the correction.

[0056] However, the third indicator icon 324 may be the same as the first indicator icon 320. Furthermore, the operation menu of the registration face list screen 300 may be used to switch between a mode in which the third indicator icon 324 is displayed and a mode in which it is not displayed. For example, in a mode in which the third indicator icon 324 is not displayed, the first indicator icon 320 and the second indicator icon 322 may be displayed, and the first indicator icon 320 may be displayed in association with the corrected second face image. In a mode in which the third indicator icon 324 is displayed, the first indicator icon 320, the second indicator icon 322, and the third indicator icon 324 may be displayed in association with the respective face images.

[0057] In another example, the display processing unit 104 may accept the selection of at least one type of face image from among the first face image, the second face image, and other face images (face images determined to not require modification) on the registration face list screen 300, and narrow down and display the selected images.

[0058] The display processing unit 104 displays the second face image after the correction process before registering it. That is, the registration face list screen 300 of FIG. 7 is displayed by the display processing unit 104 before the second face image is registered by the registration unit 106. In this case, it is preferable that the registration face list screen 300 further includes a registration button (not shown) that accepts a registration instruction. In response to accepting a press of the registration button, the registration unit 106 registers the second face image in the storage device 120 as registration authentication information. On the other hand, the registration unit 106 does not register the second face image until accepting a press of the registration button.

[0059] As a result, the image processing device 100 does not register the second face image until it receives a registration instruction, and the operator can select whether to register the second face image after checking it on the face list screen for registration 300 or to redo the correction process if necessary.

[0060] Furthermore, the display processing unit 104 may identify the similarity between the first facial image before the correction processing and the second facial image after the correction processing, and display the second facial image when the similarity meets a criterion.

[0061] The standard is that the similarity should be equal to or greater than a threshold value (e.g., 90%) to indicate that the modified second face image is suitable for authentication processing. However, if a face is modified from a profile to a frontal face, the similarity between the first face image and the second face image may be less than the threshold value. Therefore, depending on the modification content, the threshold value may be changed, or this process may be bypassed.

[0062] If the similarity does not satisfy the criteria, the display processing unit 104 may not display the second facial image, and may display a message urging the operator to redo the correction process. In this way, if the second facial image after the correction process has been performed has been changed to a person different from the first facial image, the image processing device 100 can recommend the operator to redo the correction process without registering the second facial image.

[0063] Furthermore, the display processing unit 104 may display the second face image when the stranger rejection rate of the second face image satisfies a criterion. Specifically, the criterion is that the stranger rejection rate of the second face image is equal to or greater than a threshold. Alternatively, the criterion may be that the true-person acceptance rate of the second face image is equal to or greater than a threshold. When the stranger rejection rate of the second face image is equal to or greater than the threshold, the display processing unit 104 associates the second face image with the third indicator icon 324 and displays the second face image on the registration face list screen 300 of FIG. 7 . When the stranger rejection rate of the second face image is less than the threshold, it is desirable to perform the correction process again for the second face image. Therefore, the display processing unit 104 displays the second face image on the registration face list screen 300 of FIG. 6 while associating the second face image with the second indicator icon 322. Alternatively, the display processing unit 104 may display a message recommending that the correction process be performed again (e.g., "Please try the correction process again"). The same applies to the true-person acceptance rate.

[0064] As a result, the image processing device 100 does not register, as a face image for authentication, a second face image with a low false rejection rate or a low true acceptance rate, thereby preventing a decrease in authentication accuracy.

[0065] Furthermore, the display processing unit 104 may accept changes to the correction parameters related to the correction process and repeat the correction process using the changed parameters until the above-mentioned similarity criteria or the criteria for the rejection rate of other people (or the acceptance rate of the person) are met.

[0066] Changes to the correction parameters related to the correction process may be accepted, for example, on a screen (not shown) that accepts selection of correction items and input of correction parameters, as will be described later. Alternatively, the selection of correction items and the values ​​of correction parameters may be automatically changed, the correction process may be repeated, and the display processing unit 104 may display face images that satisfy the criteria.

[0067] This allows the second facial image to meet standards suitable for authentication processing, increasing the possibility of improving the accuracy of authentication processing using the second facial image.

[0068] The display processing unit 104 may also display to what extent the second facial image satisfies the criteria. The display processing unit 104 may also display correction information that can identify a correction process recommended for satisfying the criteria.

[0069] <Operation example> The following will be explained using Fig. 2. There are various possible timings for starting the flow in Fig. 2, and examples are shown below, but the timing is not limited to these. In addition, multiple of the following can be adopted: (b1) After a new authentication system 1 is constructed and many new face images are registered (b2) When a new face image is additionally registered (b3) When a decrease in authentication accuracy is detected in an authentication system 1 that is in operation (when an erroneous authentication is found) (b4) When specified by an operator operation

[0070] The above (b2) and (b3) may be detected by periodic check by the image processing device 100. When the image processing device 100 detects at least one of the above (b2) and (b3), the image processing device 100 may notify the operator to proceed with the facial image quality confirmation process (i.e., an instruction to start the flow of FIG. 2). The notification may be, for example, outputting a pop-up message on the display of the image processing device 100, or displaying a notification icon of the application and prompting the operator to open the application to output the message.

[0071] When this flow starts, the determination unit 102 first identifies the quality of the facial image (step S101) and determines whether or not correction is required (step S103). As described above, information indicating the quality of each facial image and the determination result of whether or not correction is required are stored in the image quality information 130.

[0072] The display processing unit 104 displays the first face image determined to require retouching in association with the retouching information that can identify the recommended retouching process (step S105). As described above, the display processing unit 104 displays the registration face list screen 300 on the display of the image processing device 100.

[0073] When a decision to perform the correction process is input (YES in step S107), the registration unit 106 registers the second face image after the correction process has been performed on the first face image (step S109) as registration authentication information in the storage device 120 (step S111). Note that until a correction instruction is received in step S107, the display processing unit 104 returns to step S105 and waits for the reception of an instruction while displaying the face list screen for registration 300.

[0074] Furthermore, as shown in FIG. 8 , after the correction process is performed on the first face image in step S109, the display processing unit 104 displays the registration face list screen 300 of FIG. 7 on the display of the image processing device 100 (step S121). As described above, the registration face list screen 300 includes the second face image (no-correction face image 330) that has been subjected to the correction process. Then, when the registration unit 106 receives a press of the registration button on the registration face list screen 300 (YES in step S123), it registers the second face image (step S111 in FIG. 2 ). When the registration unit 106 does not receive a press of the registration button (NO in step S123), the process returns to step S105 in FIG. 2 , and the display processing unit 104 may wait for further correction instructions (step S107 in FIG. 2 ). However, monitoring for the press of the registration button in step S123 in FIG. 8 may be performed in parallel. That is, if the registration button is pressed while waiting for further correction instructions in step S107 of FIG. 2 (YES in step S123 of FIG. 8), the second facial image may be registered (step S111 of FIG. 2).

[0075] Furthermore, as shown in FIG. 9 , after the correction process is performed on the first facial image in step S109, the display processing unit 104 identifies the similarity between the first facial image and the second facial image and determines whether the similarity satisfies a criterion (step S131). For example, if the similarity is higher than a threshold (or the similarity may be equal to or greater than the threshold) (YES in step S131), the display processing unit 104 determines that the criterion is satisfied and displays the second facial image (step S121 in FIG. 8 ). The display processing unit 104 may display the registration face list screen 300 of FIG. 6 and display the second facial image in association with the third indicator icon 324. Then, when the registration button is pressed (YES in step S123 in FIG. 8 ), the registration unit 106 registers the second facial image (step S111 in FIG. 2 ).

[0076] On the other hand, if the similarity is lower (or equal to or lower) than the threshold value (NO in step S131), the criterion is not met, and the display processing unit 104 returns to step S105 in Fig. 2 without registering the second face image, and waits for further correction instructions (step S107 in Fig. 2). At this time, the display processing unit 104 displays the second face image in association with the second indicator icon 322 on the registration face list screen 300 in Fig. 6. Furthermore, the display processing unit 104 may display a message recommending that the second face image be corrected again.

[0077] As described above, the image processing device 100 includes the determination unit 102, the display processing unit 104, and the registration unit 106. The determination unit 102 identifies the quality of a facial image and determines whether or not retouching is necessary. The display processing unit 104 displays a first facial image determined to require retouching in association with retouching information that identifies a recommended retouching process. When a decision to perform retouching is input, the registration unit 106 registers a second facial image obtained after retouching the first facial image. As described above, according to the present embodiment, an image processing device, an image processing method, a program, and an authentication system can be obtained that can reduce the effort required to retouch a facial image used in authentication processing to make it suitable for authentication processing. Specifically, among multiple facial images for authentication, a facial image that is recommended to be retouched can be identified at a glance by the retouching information (e.g., the second indicator icon 322). Therefore, the image processing device 100 can reduce the effort required to check facial images for authentication.

[0078] Second Embodiment An image processing device 100 according to the present disclosure is the same as the image processing device 100 in Fig. 1 except that it has a configuration for accepting changes to correction content related to correction processing and performing correction processing. Since the image processing device 100 according to the present disclosure has the same configuration as the image processing device 100 in Fig. 1, it will be described using Fig. 1. Note that each element in Fig. 1 may be combined with each element in one or more embodiments to the extent that no contradiction occurs.

[0079] When the display processing unit 104 receives a selection operation of one of the plurality of first face images as a target for correction, it displays information that can identify the recommended correction process for the selected first face image.

[0080] Specifically, the description will be given using the registration face list screen 300 in Fig. 6 described in the above embodiment. In response to pressing of the second indicator icon 322 displayed in association with the first face image (the correction-requiring face image 340), the display processing unit 104 accepts the pressing as a selection operation for selecting the correction-requiring face image 340 as a correction target. The display processing unit 104 displays the correction screen 500 in Fig. 10 on the display as information that can identify the correction process recommended for the selected first face image.

[0081] Alternatively, the display processing unit 104 may display an operation menu in response to a right-click on the face image display field 310 in which the first face image (the face image requiring correction 340) is displayed on the face list screen for registration 300. The display processing unit 104 may display the correction screen 500 in response to receiving a selection of "correction process" included in the operation menu.

[0082] Furthermore, the display processing unit 104 displays a plurality of selectable correction item candidates for the correction process, and when a selection of a correction item is accepted, displays a second facial image obtained by performing the correction process on the first facial image in accordance with the accepted selection of the correction item. Specifically, the display processing unit 104 displays on the display a correction screen 500 that selectably displays a plurality of selectable correction item candidates for the correction process.

[0083] The correction process can be performed by combining various image processing applications corresponding to the correction items to satisfy the criteria. Alternatively, the correction process can be performed by using AI (artificial intelligence) to process the image to satisfy the criteria corresponding to the correction items. Here, the function for generating an image and the AI ​​are realized, for example, by a neural network. The neural network includes multiple artificial neurons and has synapses connecting each neuron. Each synapse has a weight. When such a neural network receives an input, it performs calculations using the weights associated with each synapse and produces an output according to the input. A model representing the connection relationship between neurons and synapses is stored in memory, for example, in the form of software. Alternatively, the model may be realized as a dedicated circuit. Similarly, the weights of each synapse are also stored in memory in the form of software. Alternatively, a circuit representing the weights may be implemented in a dedicated circuit. Note that when configuring the function for generating an image and the AI ​​using multiple models, all of the models do not necessarily need to be stored in the same memory. There are a variety of models using such neural networks, and a wide variety of models, such as a Transformer, a convolutional neural network (CNN), or a recurrent neural network (RNN), may be adopted or replaced to realize functions for generating images or AI.

[0084] 10 , the correction screen 500 includes an image quality display field 510, a correction reason display field 520, and a correction suggestion display field 530. The image quality display field 510 displays information indicating the quality of the first facial image. In this example, for a first image determined by the determination unit 102 to require correction, "low" is displayed, and for an image determined by the determination unit 102 to not require correction, "high" is displayed, but the information indicating the quality is not limited to this. For example, the information indicating the quality may be information obtained by scoring the quality of the image identified by the determination unit 102.

[0085] The correction reason display field 520 displays information indicating the reason why the determination unit 102 has determined that correction is necessary. In this example, "Because it is not a frontal face" is displayed as information indicating that correction is necessary because the score for the first face image being a frontal face is low. The display processing unit 104 displays information indicating the reason why the determination unit 102 has determined that correction is necessary in the correction reason display field 520 based on the evaluation result of the quality of the face image identified by the determination unit 102. For example, in the face image quality identification process by the determination unit 102 described above, the fact that the face image did not satisfy the conditions is displayed as the reason. Examples of information indicating the reason include, but are not limited to, the following: (c1) No face is included (c2) Does not look like a face (c3) Is not a frontal face (c4) Face is tilted (c5) Face is not bright enough (c6) Face is hidden (c7) Eyes are closed, mouth is open (c8) Image size is small, image resolution is low, or subject is blurred

[0086] In this example, one reason is displayed in the correction reason display field 520, but multiple reasons may be displayed in the correction reason display field 520. When multiple reasons are displayed, it is preferable to give higher priority to reasons that deviate more from the condition criteria or thresholds and display them at the top. Furthermore, when multiple conditions are not satisfied, the display processing unit 104 may display a predetermined number (e.g., three) of the reasons with the highest priority, or may display all of the reasons.

[0087] Furthermore, the display processing unit 104 may display information indicating the degree of deviation from the standard together with the reason in the correction reason display field 520. For example, this may be displayed as a percentage, level, score, or the like, such as "not a frontal face (60%)," or may be displayed as a binary value, such as "low resolution / high resolution."

[0088] The revision suggestion display field 530 displays a recommended revision suggestion for correcting the first facial image so that the first facial image satisfies the condition that caused the revision. The information displayed in the revision suggestion display field 530 is an example of information capable of identifying a recommended correction process for the first facial image. As described above, the determination unit 102 identifies recommended correction items based on the quality of the first facial image, and the display processing unit 104 displays a revision suggestion based on the identified recommended correction items in the revision suggestion display field 530. Furthermore, the revision suggestion display field 530 includes a before-correction image display field 550 and a after-correction image display field 560. The before-correction image display field 550 displays the first facial image before correction (the facial image requiring correction 340). The after-correction image display field 560 displays a candidate facial image 570 reflecting the results of the correction process for the recommended correction items.

[0089] The correction suggestion display field 530 also includes a plurality of correction item selection tabs 540. Various correction items are conceivable, and examples thereof are given below, but are not limited to these: (d1) Frontal face, correction of face tilt, etc. (d2) High resolution (d3) Brightness adjustment, contrast adjustment, blur correction, etc. (d4) Image re-registration (when no face is included, when the image size is extremely small, etc.) (d5) Aging (younger or older) (d6) Disguise function (accessories such as glasses, beard, hairstyle, hair color, makeup)

[0090] In this example, the tab for the correction item recommended as a correction suggestion (in this example, "frontal face change") is active, and the tab for the correction item for which the conditions are met and no correction is required is inactive. However, the tab for the correction item determined to be no correction required may also be displayed in a selectable state (active).

[0091] Furthermore, as described above, when correction is recommended for multiple reasons, it is preferable to display all correction items corresponding to the reasons. However, similar to the criteria for presenting reasons, correction items may also be selectively displayed. Specifically, the greater the deviation from the condition criteria or threshold, the higher the priority is set. The display processing unit 104 may then display correction items according to the following display criteria. The display criteria include any one of the correction item corresponding to the top-ranked reason, a predetermined number of correction items corresponding to a predetermined number (e.g., three) of reasons from the top rank, and correction items corresponding to all reasons.

[0092] The disguise function is a function for facial images used in cases such as facial images for tracking, and is intended to assume that the person to be tracked will disguise themselves. Therefore, for a facial image corrected by applying the disguise function, the registration unit 106 registers the corrected facial image as a new facial image of the person to be authenticated without deleting the first facial image. In this manner, multiple facial images of the person to be authenticated may be registered. However, the correction process may include a pattern for applying a disguise to the first facial image and a pattern for removing the disguise from the first facial image, assuming that the first facial image is disguised.

[0093] Furthermore, a re-correction icon 572 is displayed in association with the candidate face image 570. When the re-correction icon 572 is pressed, the display processing unit 104 automatically adjusts the correction content and updates and displays the candidate face image 570 obtained by performing the re-correction process.

[0094] The revision suggestion display field 530 further includes a revision decision button 580 and a manual adjustment button 582. When the revision decision button 580 is pressed, the registration unit 106 registers the candidate face image 570 displayed in the post-edited image display field 560 as the second face image. When the manual adjustment button 582 is pressed, the display processing unit 104 may display a parameter adjustment screen (not shown) for accepting changes to parameters for manually correcting the image.

[0095] The display processing unit 104 receives input of parameters related to the correction process, and displays a second facial image obtained by performing the correction process on the first facial image using the received input parameters.

[0096] The parameter adjustment screen includes a UI that accepts input of parameters corresponding to each correction item. After accepting the input of the parameters, it is preferable that the user can confirm the facial image after correction with the parameters applied. The display processing unit 104 may update and display the facial image after correction with each parameter input, or may display the facial image after correction with the input parameters applied after accepting the press of an operation button that instructs application of the input parameters.

[0097] Furthermore, the display processing unit 104 may display a correction screen 500 that accepts the specification of correction items, different from the correction screen 500 in Fig. 10. As shown in Fig. 11, the correction screen 500 includes a correction item setting field 700. The correction item setting field 700 includes a correction item display field 710 that displays a plurality of correction items, and check boxes 712 that accept the selection of correction items to be applied.

[0098] Of the correction items displayed in the correction item display field 710, correction items recommended by the determining unit 102 may be displayed so as to be distinguishable from other correction items. For example, the display processing unit 104 may highlight or actively display recommended correction items, and display other correction items normally or inactively. Alternatively, the display processing unit 104 may display the correction item setting field 700 with the checkboxes 712 for recommended correction items checked in advance.

[0099] The correction screen 500 further includes a candidate face image display field 720, a re-correction icon 722, and a correction confirmation button 730. The candidate face image display field 720 displays the candidate face image 570 corrected by applying the correction items selected in the correction item setting field 700. When the re-correction icon 722 is pressed, the display processing unit 104 adjusts the parameters for each correction item and displays the candidate face image 570 corrected again in the candidate face image display field 720. Note that the adjustment of the correction parameters may be performed automatically or may be configured to accept input from an operator. When the correction confirmation button 730 is pressed, the registration unit 106 registers the corrected candidate face image 570 (second face image) in the storage device 120 as registered authentication information.

[0100] As examples of parameter adjustment, the display processing unit 104 accepts settings such as changing the direction of the face, changing the resolution, changing the age of aging, etc. In addition, as examples of parameter adjustment, in the case of disguise, the display processing unit 104 accepts operations such as changing the frame of glasses, changing the length or shape of beard, or changing the hairstyle, length, or hair color.

[0101] As described above, in the image processing device 100, when a selection operation is received from among a plurality of first facial images as a target for correction, the display processing unit 104 displays information that enables identification of a recommended correction process for the selected first facial image. In this way, the image processing device 100 of this embodiment not only achieves the same effects as the above-described embodiment, but also makes it possible to grasp the content of the recommended correction process for a facial image that requires correction from among a plurality of facial images for authentication.

[0102] Furthermore, the display processing unit 104 displays a plurality of selectable correction item candidates for the correction process, and upon receiving a selection of a correction item, displays a second facial image obtained by performing the correction process on the first facial image in accordance with the received correction item. This allows the display of correction item candidates corresponding to the recommended correction process, allowing the operator to select the correction process without hesitation.

[0103] Although the embodiments of the present invention have been described above with reference to the drawings, these are merely examples of the present invention, and various other configurations can also be adopted.

[0104] (Other Embodiments) <Display of Multiple Facial Images> As described above, multiple facial images of a person to be authenticated can be registered for that person. Furthermore, both pre-edited and post-edited facial images can be stored. Therefore, the display processing unit 104 may be able to display registered or saved facial images of the person to be authenticated in association with each person to be authenticated. For example, the display processing unit 104 may display a first pre-edited facial image and a second post-edited facial image side by side. In this case, the display processing unit 104 may be able to accept, for example, an operation to return the registered facial image to the first pre-edited facial image or an operation to perform the edit process again.

[0105] Furthermore, the display processing unit 104 may display a list of multiple facial images registered as images of the person to be authenticated. In this case, since it is assumed that a facial image will be used as a tracking target, for example, the display processing unit 104 may be able to accept selection of a facial image to be used for tracking from the list.

[0106] <Simple Edit Screen Display> In the above embodiment, selecting a face image requiring edit 340 on the registration face list screen 300 displays a separate edit screen 500. In another embodiment, as shown in Fig. 12, edit processing can be realized by combining the registration face list screen 300 with a simple edit window 600. This example is particularly useful when the image processing device 100 is realized by a communication terminal with a relatively small display, such as a smartphone.

[0107] Specifically, first, on the registration face list screen 300 of FIG. 12 , the display processing unit 104 displays a fourth indicator icon 326 in the face image display field 310 in association with the correction-required face image 340 (first face image). The fourth indicator icon 326 indicates that correction is recommended for the correction-required face image 340 displayed in the face image display field 310. Furthermore, upon receiving a selection operation (such as a touch or a click) of the fourth indicator icon 326, the display processing unit 104 displays a second face image in the face image display field 310 after automatic correction processing of the first face image in accordance with the recommended correction content. If the display processing unit 104 repeatedly receives a selection operation of the fourth indicator icon 326, the display processing unit 104 can update the correction parameters to a different second face image generated by changing the correction parameters each time a selection operation is received, and display the updated second face image in the face image display field 310.

[0108] Furthermore, when the fourth indicator icon 326 is selected by long pressing or by hovering the mouse over or right-clicking, the display processing unit 104 may display a simple correction window 600 in association with the face image display field 310 in which the correction-required face image 340 is displayed. The simple correction window 600 includes a correction reason display field 610 and a correction method display field 612. The correction reason display field 610 displays the reason why the correction-required face image 340 needs to be corrected. The correction method display field 612 displays the candidate face image 570 after correction processing to correct the reason for correction, and displays a correction method for accepting correction to the candidate face image 570. For example, the correction method display field 612 displays information that indicates "click to correct to this image (candidate face image 570)." Furthermore, the display processing unit 104 may display a re-correction icon 572 in association with the candidate face image 570. When the re-correction icon 572 is pressed, the display processing unit 104 displays another candidate face image 570 generated by changing the correction parameters.

[0109] <Batch Correction Processing> In the above embodiment, a configuration has been described in which correction processing is performed for each first facial image. In other embodiments, correction processing may be performed collectively on multiple first facial images. The display processing unit 104 displays a reception unit that receives a request to perform correction processing collectively on multiple first facial images.

[0110] 13 , the display processing unit 104 may display a registration-use face batch correction screen 800 on the display. The registration-use face batch correction screen 800 includes a pre-correction image display field 810 and a post-correction image display field 820. In the pre-correction image display field 810, a plurality of face images to be registered as authentication face images (or may be already-registered face images), including a plurality of first face images, are displayed in the face image display field 310. In the post-correction image display field 820, a plurality of pre-registration face images, including a second face image after the recommended correction process, are displayed in the face image display field 310.

[0111] A correction process can be performed collectively on multiple correction-requiring facial images 340 (first facial images) displayed in the before-correction image display field 810. For example, upon receiving a press of the batch correction button 830, the display processing unit 104 displays a correction screen 500 that accepts the selection of correction items for the multiple correction-requiring facial images 340 (first facial images) and the input of correction parameters. Alternatively, the display processing unit 104 may accept a press of the batch correction button 830 after selecting at least one of the correction-requiring facial images 340 (first facial images) displayed in the before-correction image display field 810 for which correction content is to be set. Upon receiving a press of the batch correction button 830, the display processing unit 104 displays a correction screen 500 that accepts the selection of correction items for at least one selected correction-requiring facial image 340 (first facial image) and the input of correction parameters.

[0112] The display processing unit 104 displays the second face image after the correction process of the first face image has been performed in accordance with the correction items and correction parameters selected on the correction screen 500 in the corrected image display field 820. The second face image may also be further corrected individually in the corrected image display field 820. For example, by accepting a press of the re-correction icon 572 displayed in association with the candidate face image 570, the display processing unit 104 displays a new candidate face image 570 generated by changing the correction parameters of the candidate face image 570 in the face image display field 310.

[0113] Then, when the registration unit 106 receives a press of the batch registration button 832, it collectively registers the second facial images displayed in the corrected image display field 820 as facial images for authentication in the storage device 120. The batch registration button 832 is an example of the above-mentioned accepting means.

[0114] Note that the second indicator icons 322 associated with the facial images to be corrected 340 (first facial images) displayed in the before-correction image display field 810 may be individually selected. In this case, the correction screen 500 may be displayed to specify the correction details for the selected first facial images (selection of correction items and input of correction parameters). Then, upon receiving a press of the collective correction button 830, the display processing unit 104 may display, in the after-correction image display field 820, second facial images obtained after correction processing for each first facial image according to the correction details individually set for each first facial image.

[0115] <Displaying One Facial Image for Registration> In the above embodiment, the facial image list screen 300 for registration displayed a list of multiple facial images. In other embodiments, facial images may be displayed one by one. This example is particularly useful when the image processing device 100 is implemented as a communication terminal with a relatively small display, such as a smartphone. However, regardless of the size of the display, this configuration is also useful when the operator wants to enlarge and check a facial image.

[0116] 14 , the display processing unit 104 sequentially displays a plurality of face images, including the first face image, one by one on the registration face confirmation screen 400. The registration face confirmation screen 400 includes a face image display field 310 and a first indicator icon 320 or a second indicator icon 322 associated with the face image display field 310. The registration face confirmation screen 400 further includes a back button 410 and a next button 412. When the back button 410 is pressed, the display processing unit 104 displays the previous face image. When the next button 412 is pressed, the display processing unit 104 displays the next face image.

[0117] The correction process for the first face image displayed in the face image display field 310 may be realized by the content described in any of the above-mentioned embodiments. Also, the registration face confirmation screen 400 in Fig. 14 and the registration face list screen 300 in Fig. 6 may be displayed by switching between them, or may be displayed by accepting the selection of at least one of them in an operation menu for selecting the display form.

[0118] <Table-Form Display of Registration Face Images> In the above embodiment, a screen display of multiple face images, such as the registration face list screen 300, has been described. In other embodiments, the display processing unit 104 may display text information about the multiple face images recorded in a table-form data file. For example, the table-form data file may associate, for each of the multiple face images, the determination result of whether correction is necessary, information about the reason if correction is necessary, and text information indicating whether correction has been made. Furthermore, the table-form data file may also include path information where the face images are stored. The path is linked to a storage location, and by selecting the link, the display processing unit 104 may display the corresponding face image in a separate window. This may allow the operator to view the face images.

[0119] Another example of the present disclosure may be a program causing at least one computer to execute the method disclosed above, or a computer-readable recording medium having such a program recorded thereon. This recording medium includes a non-transitory tangible medium. The computer program includes computer program code that, when executed by a computer, causes the computer to perform the image processing method on an image processing device.

[0120] Any combination of the above components, and conversion of the present disclosure into a method, device, system, recording medium, computer program, etc., are also valid aspects of the present disclosure.

[0121] Furthermore, the various components of the present disclosure do not necessarily have to be independent entities, but may be formed as a single member by multiple components, one component may be formed from multiple components, one component may be a part of another component, or part of one component may overlap with part of another component, etc.

[0122] Furthermore, although the methods and computer programs disclosed herein describe a number of steps in a sequential order, the order in which the steps are performed does not limit the order in which the steps are performed. Therefore, when implementing the methods and computer programs disclosed herein, the order of the steps can be changed as long as it does not cause any problems in terms of the content.

[0123] Furthermore, the steps of the method and computer program disclosed herein are not limited to being executed at different times, and may be executed while one step is being executed, or may be executed partially or entirely overlapping with another step, etc.

[0124] Although the present disclosure has been described above with reference to the embodiments, the present disclosure is not limited to the above-described embodiments. Various modifications that can be understood by those skilled in the art can be made to the configuration and details of the present disclosure within the scope of the present disclosure. Furthermore, each embodiment can be combined with other embodiments as appropriate. Note that, when information about a user (authentication target) is acquired and used in the present disclosure, this must be done lawfully.

[0125] In addition, although the flowcharts used in the above description show a sequence of steps (processes), the order of steps executed in each embodiment is not limited to the sequence shown in the flowcharts. In each embodiment, the order of steps shown in the diagrams can be changed as long as it does not cause any problems in terms of the content.

[0126] Some or all of the above embodiments may be described as in the following supplementary notes, but are not limited to them. 1. An image processing device comprising: a determination means for identifying the quality of a facial image and determining whether retouching is necessary; a display processing means for displaying a first facial image determined to require retouching in association with retouching information capable of identifying a recommended retouching process; and a registration means for registering a second facial image after the retouching process has been performed on the first facial image when a decision to perform the retouching process is input. 2. An image processing device described in 1., wherein the display processing means displays the second facial image after the retouching process has been performed before registering it. 3. An image processing device described in 1. or 2., wherein the display processing means identifies a similarity between the first facial image before the retouching process and the second facial image after the retouching process has been performed, and displays the second facial image when the similarity meets a criterion. 4. An image processing device described in 1. to 3. 5. The image processing device described in any one of 1. to 4., wherein the display processing means displays at least two of the first facial images on the same screen. 5. The image processing device described in any one of 1. to 4., wherein the display processing means accepts input of parameters related to the correction process, and displays the second facial image obtained by performing the correction process on the first facial image using the accepted input parameters. 6. The image processing device described in any one of 1. to 5., wherein the display processing means selectably displays candidates for a plurality of correction items for the correction process, and, upon accepting selection of the correction item, displays the second facial image obtained by performing the correction process on the first facial image in accordance with the accepted selection of the correction item. 7. The image processing device described in any one of 1. to 6., wherein the facial image is a facial image used in authentication processing and is registered in a storage unit. 8. The image processing device described in 3., wherein the display processing means displays the second facial image when the false rejection rate of the second facial image satisfies a standard.9. The image processing device described in 8. which cites 3. or 3., wherein the display processing means accepts changes to parameters related to the correction process and repeats the correction process using the changed parameters until the criterion is met. 10. The image processing device described in 4., wherein the display processing means, upon accepting a selection operation of one of the plurality of first face images as a correction target, displays information capable of identifying the recommended correction process for the selected first face image. 11. The image processing device described in 10. which cites 4. or 4., wherein the display processing means displays a receiving means for receiving a request operation to perform the correction process collectively on the plurality of first face images.

[0127] 12. An authentication system comprising: an authentication device that performs authentication processing using facial images; and an image processing device that processes the facial images used by the authentication device for the authentication processing, wherein the image processing device has: a determination means that identifies the quality of the facial image and determines whether retouching is necessary; a display processing means that displays a first facial image determined to require retouching in association with retouching information that can identify a recommended retouching process; and a registration means that, when a decision input is made to perform the retouching process, registers a second facial image after the retouching process has been performed on the first facial image. 13. An image processing method in which one or more computers identify the quality of a facial image and determine whether retouching is necessary; display the first facial image determined to require retouching in association with retouching information that can identify a recommended retouching process; and, when a decision input is made to perform the retouching process, registers the second facial image after the retouching process has been performed on the first facial image. 14. 15. A computer-readable recording medium storing a program for causing a computer to execute: a determination process for identifying the quality of a facial image and determining whether or not retouching is necessary; a display process for displaying a first facial image determined to require retouching in association with retouching information that can identify a recommended retouching process; and a registration process for registering a second facial image after the retouching process has been performed on the first facial image, when a decision input is made to perform the retouching process. 15. A computer-readable recording medium storing a program for causing a computer to execute: a determination process for identifying the quality of a facial image and determining whether retouching is necessary; a display process for displaying a first facial image determined to require retouching in association with retouching information that can identify a recommended retouching process; and a registration process for registering a second facial image after the retouching process has been performed on the first facial image, when a decision input is made to perform the retouching process.

[0128] Furthermore, some or all of the configurations described in Supplementary Notes 2 to 11 that are subordinate to the above-mentioned Supplementary Note 1 (image processing device) may also be subordinate to the authentication system of Supplementary Note 12, the image processing method of Supplementary Note 13, the program of Supplementary Note 14, and the recording medium of Supplementary Note 15 in the same subordinate relationship as Supplementary Notes 2 to 11. Furthermore, not limited to Supplementary Notes 12, 13, 14, and 15, some or all of the configurations described as Supplements may be subordinate to various hardware, software, various recording means for recording software, or systems, within the scope of each of the above-mentioned embodiments.

[0129] This application claims priority based on Japanese Patent Application No. 2024-021393, filed February 15, 2024, the disclosure of which is incorporated herein in its entirety by reference.

[0130] REFERENCE SIGNS LIST 1 Authentication system 100 Image processing device 102 Determination unit 104 Display processing unit 106 Registration unit 120 Storage device 130 Image quality information 140 Image information 200 Authentication device 300 Registration face list screen 400 Registration face confirmation screen 500 Correction screen 600 Simple correction window 800 Registration face batch correction screen 1000 Computer 1010 Bus 1020 Processor 1030 Memory 1040 Storage device 1050 Input / output interface 1060 Network interface

Claims

1. An image processing device comprising: a judgment means for identifying the quality of a facial image and determining whether or not correction is necessary; a display processing means for displaying a first facial image that has been determined to require correction in association with correction information that can identify a recommended correction process; and a registration means for registering a second facial image after the correction process has been performed on the first facial image when a decision to perform the correction process is input.

2. An image processing device according to claim 1, wherein the display processing means displays the second face image after the correction processing has been performed before registering it.

3. An image processing device according to claim 1 or 2, wherein the display processing means determines the degree of similarity between the first facial image before the correction processing and the second facial image after the correction processing, and displays the second facial image when the degree of similarity satisfies a criterion.

4. An image processing device according to any one of claims 1 to 3, wherein the display processing means displays at least two of the first facial images on the same screen.

5. An image processing device according to any one of claims 1 to 4, wherein the display processing means accepts input of parameters related to the correction process, and displays the second facial image obtained by performing the correction process on the first facial image using the accepted input parameters.

6. An image processing device according to any one of claims 1 to 5, wherein the display processing means displays a plurality of candidate correction items for the correction process in a selectable manner, and when a selection of a correction item is accepted, displays the second facial image obtained by performing the correction process on the first facial image in accordance with the correction item for which the selection has been accepted.

7. An image processing device according to any one of claims 1 to 6, wherein the face image is a face image used for authentication processing and is registered in a storage unit.

8. An image processing device according to claim 3, wherein the display processing means displays the second face image when the false rejection rate of the second face image satisfies a criterion.

9. An image processing device according to claim 3 or claim 8 which cites claim 3, wherein the display processing means accepts changes to parameters relating to the correction process and repeats the correction process using the changed parameters until the criteria are met.

10. An image processing device according to claim 4, wherein the display processing means, when accepting a selection operation of one of the plurality of first facial images as an object to be corrected, displays information that can identify the correction processing recommended for the selected first facial image.

11. An image processing device according to claim 4 or claim 10 which cites claim 4, wherein the display processing means displays a reception means which receives a request operation to collectively perform the correction processing on a plurality of the first facial images.

12. An authentication system comprising: an authentication device that performs authentication processing using a facial image; and an image processing device that processes the facial image used by the authentication device for the authentication processing, wherein the image processing device has: a determination means that identifies the quality of the facial image and determines whether or not correction is necessary; a display processing means that displays a first facial image that has been determined to require correction in association with correction information that can identify a recommended correction process; and a registration means that, when a decision input is made to perform the correction process, registers a second facial image after the correction process has been performed on the first facial image.

13. An image processing method in which one or more computers identify the quality of a facial image, determine whether or not correction is necessary, display a first facial image determined to require correction in association with correction information that can identify recommended correction processing, and, when a decision to perform the correction processing is input, register a second facial image after the correction processing has been performed on the first facial image.

14. A computer-readable recording medium storing a program for causing a computer to execute: a determination process for identifying the quality of a facial image and determining whether or not correction is necessary; a display process for displaying a first facial image determined to require correction in association with correction information that can identify the recommended correction process; and a registration process for registering a second facial image after the correction process has been performed on the first facial image when a decision to perform the correction process is input.

Images