JP7168192B2 - Face image adequacy determination device, face image adequacy determination method, program, and recording medium - Google Patents

Description

The present invention relates to a face image adequacy determination device, a face image adequacy determination method, a program, and a recording medium.

2. Description of the Related Art In various fields, personal authentication by facial image authentication using facial images has become common. In the face image authentication, usually, as a master image, a face image such as an identification photograph of the person is first stored in a database, and then the face image of the subject obtained live is collated as a query image to perform authentication. is done.

The said ID photograph generally requires a specific style in terms of size, background, and the like. For this reason, there is known a device for determining whether a face image is suitable for use as an ID photograph based on the background in the face image, the number of faces, their orientations, their sizes, and the like (Patent Document 1).

JP 2007-034804 A

However, there are cases where the face image stored as the master image cannot be used for matching. For example, if the master image is a blurred photograph, authentication may not be possible even if the subject is the person himself/herself. For this reason, the current situation is that a camera-shaky photograph is not used as a master image. However, for unknown reasons, there are face images that cannot be used as master images, even if they are not blurry pictures, and it is difficult to determine which face images can be used as master images.

SUMMARY OF THE INVENTION Accordingly, it is an object of the present invention to provide a new system for determining whether facial images are inappropriate for facial image authentication.

In order to achieve the above object, the face image adequacy determination device of the present invention comprises:
a face image acquisition unit that acquires a face image;
a binarization processing unit that generates a binarized image from the face image based on a predetermined threshold;
a part detection unit for detecting a face contour, a right eye area including the right eye, a left eye area including the left eye, a nose area including the nose, and a mouth area including the mouth from the face image;
a first straight line, in the binarized image, having a first start point at any position of the boundary of the right eye region and a first end point at an intersection with the contour of the right face below the first start point; and a detection line setting unit for setting a second straight line having a second start point at any position of the boundary of the left eye region and a second end point at an intersection with the contour of the left face below the second start point;
a detection line setting unit configured to detect changes in brightness of the first straight line and the second straight line except for the right eye region, the left eye region, the nose region, and the mouth region;
and a determination unit that determines that the face image is an image inappropriate for face image authentication when the luminance change is detected for at least one of the first straight line and the second straight line. do.

The face image adequacy determination method of the present invention includes:
a binarization processing step of generating a binarized image from the face image based on a predetermined threshold;
a part detection step of detecting a face contour, a right eye region including the right eye, a left eye region including the left eye, a nose region including the nose, and a mouth region including the mouth from the face image;
a first straight line, in the binarized image, having a first start point at any position of the boundary of the right eye region and a first end point at an intersection with the contour of the right face below the first start point; and a detection line determining step of setting a second straight line having a second start point at any position of the boundary of the left eye region and a second end point at an intersection with the contour of the left face below the second start point;
a luminance change detection step of detecting a luminance change of the first straight line and the second straight line, excluding the right eye region, the left eye region, the nose region and the mouth region;
and a determination step of determining that the face image is an image inappropriate for face image authentication when the luminance change is detected for at least one of the first straight line and the second straight line. do.

The face image authentication device of the present invention is
With respect to the candidate face image of the master image, if the luminance change is detected, the candidate face image is determined to be an image inappropriate for the face image authentication, and if the luminance change is not detected, the candidate face image is determined. a suitability determination unit that determines that the face image is suitable for face image authentication;
a storage unit that stores the candidate face image of the master image as a master image when the candidate face image is determined to be an appropriate image;
a face image authentication unit that compares the query image and the master image and authenticates when they match,
The suitability determination unit is the suitability determination device of the present invention.

The face image authentication method of the present invention is
With respect to the candidate face image of the master image, if the luminance change is detected, the candidate face image is determined to be an image inappropriate for the face image authentication, and if the luminance change is not detected, the candidate face image is determined. a suitability determination step of determining that the face image is an image suitable for face image authentication;
a storage step of storing the candidate face image of the master image as a master image when the candidate face image is determined to be an appropriate image;
A face image authentication step of matching the query image and the master image and authenticating if they match,
The adequacy determination step is the adequacy determination method of the present invention.

A program according to the present invention causes a computer to execute the face image adequacy determination method according to the present invention.

A program of the present invention causes a computer to execute the face image authentication method of the present invention.

A computer-readable recording medium of the present invention is characterized by recording the program of the present invention.

According to the present invention, for example, a face image inappropriate for image authentication can be easily determined regardless of whether or not the photograph is blurred.

1 is a block diagram showing an example of a suitability determination device according to Embodiment 1; FIG. 2 is a block diagram showing an example of the hardware configuration of the adequacy determination device of Embodiment 1; FIG. 4 is a flow chart showing an example of a suitability determination method according to Embodiment 1. FIG. 1 is a schematic diagram showing an example of image processing by the adequacy determination device of Embodiment 1, (A) is a binarized image, (B) is a face image (grayscale image) before binarization, (C) shows the mouth region and the like in the binarized image, and shows the first straight line and the second straight line; (D) shows (C) and the second straight line; 3 straight lines are shown respectively.

As a result of intensive research, the present inventors have found that, in facial image authentication, authentication becomes difficult if there is a shadow in the region below the eyes of the face image. Based on this finding, the inventors have established the present invention capable of determining whether a shadow is generated in such an area and determining whether or not a face image is suitable for use in face image authentication.

An embodiment of the present invention will be described with reference to the drawings. The invention is not limited to the following embodiments. In each figure below, the same reference numerals are given to the same parts. In addition, the description of each embodiment can be used with reference to each other's description unless otherwise specified, and the configurations of the respective embodiments can be combined unless otherwise specified. In the present invention, the directions in the face image and the binarized image are upward on the head side, downward on the neck side, rightward on the right eye side, and leftward on the left eye side. In this embodiment, a "face image" is an image before binarization processing, and a "binarized image" is an image obtained by binarizing the face image.

[Embodiment 1]
An example of the suitability determination device and the suitability determination method of the present invention will be described with reference to the drawings.

FIG. 1 is a block diagram showing the configuration of an example of a facial image adequacy determination apparatus 1 according to Embodiment 1 of the present invention. Referring to FIG. 1 , the suitability determination device 1 has a face image acquisition section 10 , a binarization processing section 11 , a part detection section 12 , a detection line setting section 13 , a luminance change detection section 14 and a determination section 15 . The suitability determination device 1 is also referred to as, for example, a face image suitability determination system. The suitability determination device 1 is not limited to the configuration described above, and may have, for example, a storage unit, an output unit, and the like.

The suitability determination device 1 may be, for example, a single suitability determination device including the above units, or may be a suitability determination device to which the above units can be connected via a communication network. The communication line network is not particularly limited, and a known network can be used, and may be wired or wireless, for example. The communication network includes, for example, the Internet line, telephone line, LAN (Local Area Network), WiFi (Wireless Fidelity), and the like.

FIG. 2 illustrates a block diagram of the hardware configuration of the adequacy determination device 1 of FIG. Referring to FIG. 2, the suitability determination device 1 includes, for example, a CPU (Central Processing Unit) 101, a memory 102, an interface (I/F) 103, an input device 104, a display 105, a communication device 106, a storage device 107, and the like. . Each unit of the adequacy determination device 1 is connected via an I/F 103, for example. In the hardware configuration, internal communication between circuits is connected by a bus.

The CPU 101 is responsible for overall control of the suitability determination device 1 . In the adequacy determination device 1, the CPU 101 executes, for example, the program 108 of the present invention and other programs, and reads and writes various information. Specifically, in the suitability determination device 1, for example, the CPU 101 functions as a face image acquisition unit 10, a binarization processing unit 11, a part detection unit 12, a detection line setting unit 13, a luminance change detection unit 14, and a determination unit 15. do.

The I/F 103 connects the functional units such as the CPU 101 and the memory 102 described above, for example. Also, the I/F 103 can be connected to an external device such as a camera, for example. The adequacy determination device 1 can be connected to the communication network by the communication device 106 connected to the I/F 103, and can also be connected to the external device via the communication network. When the adequacy determination device 1 has the output section, for example, the I/F 103 or the communication device 106 connected to the I/F 103 functions as the output section.

The memory 102 includes, for example, a main memory, which is also referred to as main memory. When the CPU 101 performs processing, the memory 102 reads various operating programs 108 such as the program of the present invention stored in an auxiliary storage device described later, and the CPU 101 receives data from the memory 102. , to run the program. The main memory is, for example, RAM (random access memory). Memory 102, for example, further includes a ROM (Read Only Memory).

The storage device 107 is also called a so-called auxiliary storage device, for example, in contrast to the main memory (main storage device). The storage device 107 includes, for example, a storage medium and a drive that reads from and writes to the storage medium. The storage medium is not particularly limited, and may be, for example, a built-in type or an external type. A DVD, a flash memory, a memory card, etc. can be cited, and the drive is not particularly limited. Examples of the storage device 107 include a hard disk drive (HDD) in which a storage medium and a drive are integrated, a solid state drive (SSD), and the like. The storage device 107 stores an operation program such as the program 108 of the present invention, for example. The storage device 107 is, for example, the storage unit.

The adequacy determination device 1 further optionally has an input device 104 and a display 105, for example. The input device 104 is, for example, a touch panel, keyboard, mouse, or the like. The display 105 is, for example, an LED display, a liquid crystal display, or the like.

The face image acquisition unit 10 acquires a face image. The face image may be, for example, a face image used as a master image (an image to be registered in advance) or a face image used as a query image (an image to be matched against the master image) in face image authentication. good. The face image acquisition unit 10 may be, for example, an imaging unit that itself captures a face image of a subject, or an input unit that inputs a captured face image. When the face image acquiring unit 10 is the imaging unit, for example, the camera can be used. When the face image acquisition unit 10 is the input unit, the suitability determination device 1 may be electrically connected to the camera, and receive face image information captured by the camera. The connection between the adequacy determination device 1 and the camera is not particularly limited, and may be wired or wireless, for example. The adequacy determination device 1 and the camera can be connected via, for example, a communication network, and the communication network is not particularly limited, and is, for example, as described above.

The binarization processing unit 11 generates a binarized image from the face image by binarization processing based on a predetermined threshold value. The binarization process is a process of modifying an image into an image (binarized image) having only two pixel values. The threshold value can be appropriately determined depending on, for example, whether face image adequacy determination is performed relatively strictly or relatively loosely. The threshold can be set by, for example, using a pixel value indicating brightness in an image as an arbitrary reference value. The pixel value is, for example, luminance. Generally, a higher value of the luminance indicates a brighter color, and a lower value indicates a darker color. In the binarization process, for example, each pixel in the image is converted to 1 pixel (white pixel) if the pixel value exceeds the threshold (or is equal to or greater than the threshold), and if the pixel value is equal to or less than the threshold, If (or less than a threshold), convert to 0 pixels (black pixels). In addition, in the binarization process, for example, black and white may be reversed.

The part detection unit 12 detects the outline of the face, the right eye region including the right eye, the left eye region including the left eye, the nose region including the nose, and the mouth region including the mouth from the face image that has not been binarized. . Each of the above regions may include the right eye, left eye, nose, or mouth, and the shape thereof is not particularly limited, and may be, for example, circular or angular. Specifically, for example, each of the regions (right eye region, left eye region, nose region, and mouth region) has the maximum length V in the vertical direction for each part in the face image. A quadrangle with a horizontal maximum length H in the vertical left-right direction, a circle with a diameter equal to the longer of the maximum lengths V and H, or a longer one of the maximum lengths V and H and a shorter one It can be elliptical with a minor axis. The detection of the contour of the face and the respective parts by the part detection unit 12 can utilize known part detection means such as OpenCV.

In the binarized image, the detection line setting unit 13 sets any position of the boundary of the right eye region as a first start point, and the intersection with the contour of the right face below the first start point as a first end point. and a second straight line whose second start point is any position of the boundary of the left eye region and whose second end point is the intersection with the contour of the left face below the second start point do. The first straight line may be a straight line in the above-described direction, and a specific example is a perpendicular line from the first starting point. The second straight line may be a straight line in the direction described above, and a specific example is a perpendicular line from the second starting point. The perpendicular line is, for example, a line passing through cheek regions on the right and left faces. If there is a change in brightness in the cheek region, that is, if there is a shadow, it is highly likely that the face image cannot be used for face image authentication. For this reason, it is preferable to set the perpendicular line as a detection line for luminance change, which will be described later. The perpendicular line is, for example, a line perpendicular to a line connecting the right eye and the left eye. The line connecting the right eye and the left eye is, for example, a line connecting the center of the right eye area and the center of the left eye area.

A luminance change detection unit 14 detects luminance changes of the first straight line and the second straight line except for the right eye area, the left eye area, the nose area, and the mouth area. The change in luminance means, for example, the presence of two values on the first straight line or the second straight line, specifically, for example, the presence of 0 pixels (black pixels). means

The determination unit 15 determines that the face image is an image inappropriate for face image authentication when the luminance change is detected for at least one of the first straight line and the second straight line.

For example, the detection line setting unit 13 may further set a third straight line connecting the midpoint of the first straight line and the midpoint of the second straight line. When the third straight line is set, the brightness change detection unit 14 also detects brightness changes for the third straight line, for example, except for the right eye region, the left eye region, the nose region, and the mouth region. . Then, for example, the determination unit 15 further determines that the face image is an image inappropriate for face image authentication, even when the luminance change is detected for the third straight line. That is, the determination unit 15 determines that the face image is inappropriate for face image authentication when the luminance change is detected in any one of the first straight line, the second straight line, and the third straight line. image.

Next, the suitability determination method of this embodiment will be described. The suitability determination method of the present embodiment can be implemented, for example, by using the face image suitability determination apparatus 1 shown in FIGS. Note that the suitability determination method of the present embodiment is not limited to the use of the suitability determination device 1 . Also, the description of the suitability determination method of the present embodiment can be incorporated into the suitability determination device 1 described above.

The binarization processing step generates a binarized image from the acquired face image based on a predetermined threshold value. The binarization processing step can be executed by the binarization processing section 11 of the adequacy determination device 1, for example. Acquisition of the face image can be performed by the face image acquisition unit 10 of the adequacy determination device 1, for example.

The part detection step detects a face contour, a right eye region including the right eye, a left eye region including the left eye, a nose region including the nose, and a mouth region including the mouth from the face image. The site detection step can be executed by the site detection unit 12 of the adequacy determination device 1, for example.

The execution order of the binarization processing step and the site detection step is not particularly limited. may be executed to In the latter case, the order of execution is not particularly limited and is arbitrary.

In the detection line setting step, in the binarized image, any position of the boundary of the right eye region is set as a first start point, and an intersection with the contour of the right face downward from the first start point is set as a first end point. and a second straight line whose second start point is any position of the boundary of the left eye region and whose second end point is the intersection with the contour of the left face below the second start point do. The detection line setting step can be executed by the detection line setting unit 13 of the adequacy determination device 1, for example. The detection line setting step may, for example, further set a third straight line connecting the midpoint of the first straight line and the midpoint of the second straight line. The first straight line and the second straight line are, for example, perpendicular lines as described above.

The luminance change detection step detects luminance changes with respect to the first straight line and the second straight line, except for the right eye area, the left eye area, the nose area, and the mouth area. The brightness change detection step can be executed by the brightness change detection unit 14 of the suitability determination device 1, for example. When the third straight line is set in the detection line setting step, the luminance change detecting step further includes, for example, the third straight line as well, except for the right eye region, the left eye region, the nose region, and the mouth region. , to detect luminance changes. The order of detection of luminance changes in the first straight line, the second straight line, and optionally the third straight line is not particularly limited, and may be performed simultaneously or separately, for example. In the latter case, the order of execution is not particularly limited and is arbitrary.

The determining step determines that the face image is an image inappropriate for face image authentication when the luminance change is detected for at least one of the first straight line and the second straight line. The judging step can be executed by the judging section 15 of the suitability judging device 1, for example. When the third straight line is set, the determining step further determines that the face image is an image inappropriate for face image authentication even when the luminance change is detected for the third straight line. judge. On the other hand, in the determination step, for example, when no luminance change is detected in any of the first straight line, the second straight line, and optionally the third straight line, the face image is used for face image authentication. It may be determined that the image is appropriate.

Next, the suitability determination method according to Embodiment 1 of the present invention will be described more specifically with reference to the flow chart of FIG. 3 and FIG. Note that the suitability determination method of the present invention is not limited to these examples. FIG. 4 is a schematic diagram showing an example of image processing in the adequacy determination device and the adequacy determination method according to Embodiment 1 of the present invention.

First, a face image is acquired (S100). The face image is, for example, a color image or a grayscale image.

Then, as the binarization processing step, as shown in FIG. 4A, the binarization processing unit 11 generates a binarized image 20 from the acquired face image based on a predetermined threshold value ( S101).

As the part detection step, as shown in FIG. 4B, the part detection unit 12 detects a facial contour 121, a right eye region 122a including the right eye, and a left eye region including the left eye for the face image 30 before binarization processing. 122b, the nose region 123 including the nose, and the mouth region 124 including the mouth are detected (S102).

As the detection line setting step, as shown in FIG. 4C, the detection line setting unit 13 sets the lower boundary of the right eye region 122a in the binarized image 20 as the first start point, and the line below the first start point. A first straight line 131 whose first end point is the point of intersection with the contour of the right face in the direction of A second straight line 132 is set as a second end point. A first straight line 131 and a second straight line 132 are perpendicular lines connecting the first starting point or the second starting point and the first end point or the second end point, respectively. Furthermore, as shown in FIG. 4D, a third straight line 133 is set that connects the midpoint of the first straight line 131 and the midpoint of the second straight line 132 (S103).

In the detection line setting step, for example, the binarized image 20 and the face image 30 before binarization processing are compared, and the outline 121, the right eye region 122a, the left eye region 122b, and the nose region detected in the face image 30 are detected. 123 , and mouth region 124 may be set in the binarized image 20 . As a result, the first straight line 131, the second straight line 132, and optionally the third straight line 133 can be set as described above.

In the brightness change detection step, the brightness change detection unit 14 detects the brightness of the first straight line 131, the second straight line 132, and the third straight line 133 except for the right eye region 122a, the left eye region 122b, the nose region 123, and the mouth region . A change is detected (S104).

Then, as the determination step, first, it is determined whether or not there is a luminance change in the first straight line 131 (S105-1). If there is a luminance change in the first straight line 131 (YES), the determination unit 15 determines that the facial image is inappropriate for facial image authentication (S106), and ends the process (END). On the other hand, if there is no luminance change on the first straight line 131 (NO), then it is determined whether there is a luminance change on the second straight line 132 (S105-2). If there is a luminance change in the second straight line 132 (YES), the determination unit 15 determines that the facial image is inappropriate for facial image authentication (S106), and ends the process (END). On the other hand, if there is no luminance change on the second straight line 132 (NO), it is further determined whether there is a luminance change on the third straight line 133 (S105-3). If there is a luminance change in the third straight line 133 (YES), the determination unit 15 determines that the facial image is inappropriate for facial image authentication (S106), and ends the process (END). On the other hand, if there is no change in luminance on the third straight line 133 (NO), that is, if there is no change in luminance on any of the first straight line 131, the second straight line 132, and the third straight line 133, the determination unit 15 determines that the face The image is determined to be suitable for face image authentication (S107), and the process ends (END).

Note that the order of execution of the determination steps (the S105-1 step, the S105-2 step, and the S105-3 step) shown in FIG. 3 is an example and is not particularly limited. The S105-1 step, the S105-2 step, and the S105-3 step may be performed simultaneously or separately, for example. In the latter case, the order of execution is not particularly limited and is arbitrary. In this case, for example, if there is a luminance change in any one of the steps S105-1, S105-2, and S105-3, which are executed simultaneously or in any order, the face image is a face If the image is determined to be inappropriate for image authentication and there is no luminance change in any of the steps S105-1, S105-2, and S105-3, the face image is an image suitable for face image authentication. Judge as.

In order to detect the brightness change of the face image that has undergone the binarization process as described above, the adequacy determination device 1 detects, for example, the position of the eyes, mouth and nose in the face image, excluding the positions of the eyes, mouth and nose. can determine whether a shadow exists in the region of . Thereby, for example, a face image inappropriate for face image authentication can be easily determined.

[Embodiment 2]
A program according to Embodiment 2 of the present invention is a program capable of executing the adequacy determination method of the present invention on a computer. Alternatively, the program according to Embodiment 2 of the present invention may be recorded on a computer-readable recording medium, for example. The recording medium is not particularly limited, and includes, for example, the above-described storage medium.

[Embodiment 3]
A face image authentication device and a face image authentication method of the present invention will be described.

The face image authentication device according to the third embodiment of the present invention has a face image adequacy determination unit, a storage unit, and a face image authentication unit. The face image authentication device may be, for example, one face image authentication device including the above units, or may be a face image authentication device to which the above units can be connected via a communication network. The communication network is not particularly limited, and is, for example, as described above. The face image authentication device is also called a face image authentication system, for example. The face image authentication device is not limited to the configuration described above, and may have, for example, an output unit. The face image authentication device may have, for example, a configuration in which the adequacy determination device 1 of the first embodiment further includes the storage unit and the face image authentication unit.

The suitability determination unit determines that the candidate face image is an image inappropriate for the face image authentication when the luminance change is detected in the candidate face image of the master image (preliminarily registered image). If the luminance change is not detected, the candidate face image is determined to be suitable for face image authentication. For example, the suitability determination unit further disqualifies the candidate face image from the face image authentication when the luminance change is detected in the candidate face image of the query image (the image to be matched against the master image). If the candidate face image is determined to be an appropriate image and the luminance change is not detected, the candidate face image may be determined to be an image appropriate for face image authentication. When it is determined that the candidate face image of the query image is an appropriate image, the face image authentication unit, which will be described later, uses the candidate face image of the query image as the query image, for example. The suitability determination unit is, for example, the suitability determination device 1 of the first embodiment.

The storage unit stores the candidate face image of the master image as a master image when it is determined that the candidate face image of the master image is an appropriate image. As the storage unit, for example, a storage device as described above can be used.

The face image authentication unit compares the query image and the master image, and authenticates when they match. The face image authentication unit may, for example, check whether the query image and the master image match, or check whether the query image and the master image are similar. The match and similarity can be determined, for example, based on similarity obtained by a known similarity calculation between the query image and the master image. The matching method by the face image authentication unit is not particularly limited. A conventionally known face authentication method can be used, such as a method using a feature amount such as the direction of a line connecting the . The face image authentication section is executed by, for example, a CPU.

Next, a facial image authentication method according to Embodiment 3 of the present invention will be described. The face image authentication method according to Embodiment 3 of the present invention can be implemented using, for example, the face image authentication device according to Embodiment 3 of the present invention. The face image authentication method according to Embodiment 3 of the present invention is not limited to use of the face image authentication device. Also, the description of the face image authentication method according to the third embodiment of the present invention can be applied to the face image authentication device described above.

A facial image authentication method according to Embodiment 3 of the present invention includes a facial image adequacy determination step, a storage step, and a facial image authentication step.

The suitability determination step determines that the candidate face image of the master image is an image inappropriate for face image authentication when the luminance change is detected in the candidate face image of the master image, and determines that the luminance change is detected. If not, it is determined that the candidate face image is suitable for face image authentication. For example, in the suitability determination step, when the luminance change is detected in the candidate face image of the query image, the candidate face image is determined to be an image inappropriate for the face image authentication, and the luminance If no change is detected, the candidate face image may be determined to be an image suitable for face image authentication. When the candidate face image of the query image is determined to be an appropriate image, the candidate face image of the query image is used as the query image in the face image authentication process described later, for example. The adequacy determination step is, for example, the adequacy determination method of the present invention. The suitability determination step can be executed, for example, by the suitability determination unit of the face image authentication device.

The storing step stores the candidate face image of the master image as a master image when it is determined that the candidate face image of the master image is an appropriate image. The storage step can be executed by, for example, the storage unit of the face image authentication device.

The face image authentication step compares the query image and the master image, and authenticates when they match. The face image authentication step can be executed, for example, by the face image authentication section of the face image authentication device.

[Embodiment 4]
A program according to Embodiment 4 of the present invention is a program capable of executing the face image authentication method of the present invention on a computer. Alternatively, the program according to Embodiment 4 of the present invention may be recorded on a computer-readable recording medium, for example. The recording medium is not particularly limited, and includes, for example, the above-described storage medium.

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

<Appendix>
Some or all of the above-described embodiments and examples can be described as in the following appendices, but are not limited to the following.
(Appendix 1)
a face image acquisition unit that acquires a face image;
a binarization processing unit that generates a binarized image from the face image based on a predetermined threshold;
a part detection unit for detecting a face contour, a right eye area including the right eye, a left eye area including the left eye, a nose area including the nose, and a mouth area including the mouth from the face image;
a first straight line, in the binarized image, having a first start point at any position of the boundary of the right eye region and a first end point at an intersection with the contour of the right face below the first start point; and a detection line setting unit for setting a second straight line having a second start point at any position of the boundary of the left eye region and a second end point at an intersection with the contour of the left face below the second start point;
a detection line setting unit configured to detect changes in brightness of the first straight line and the second straight line except for the right eye region, the left eye region, the nose region, and the mouth region;
and a determination unit that determines that the face image is an image inappropriate for face image authentication when the luminance change is detected for at least one of the first straight line and the second straight line. Appropriateness determination device for facial images.
(Appendix 2)
Said first straight line is a perpendicular line from said first starting point, and said second straight line is a perpendicular line from said second starting point.
(Appendix 3)
Further, the detection line setting unit sets a third straight line connecting the midpoint of the first straight line and the midpoint of the second straight line,
the brightness change detection unit detects brightness changes for the third straight line except for the right eye region, the left eye region, the nose region and the mouth region;
Supplementary note 1 or 2, wherein the determination unit determines that the face image is an image inappropriate for face image authentication even when the luminance change is detected for the third straight line. Appropriate judgment device.
(Appendix 4)
With respect to the candidate face image of the master image, if the luminance change is detected, the candidate face image is determined to be an image inappropriate for the face image authentication, and if the luminance change is not detected, the candidate face image is determined. a suitability determination unit that determines that the face image is suitable for face image authentication;
a storage unit that stores the candidate face image of the master image as a master image when the candidate face image is determined to be an appropriate image;
a face image authentication unit that compares the query image and the master image and authenticates when they match,
A face image authentication device, wherein the suitability determination unit is the suitability determination device according to any one of Appendices 1 to 3.
(Appendix 5)
The suitability determination unit determines that the candidate face image is inappropriate for face image authentication when the luminance change is detected in the candidate face image of the query image, and determines that the luminance change is detected. If not, determine that the candidate face image is an image suitable for face image authentication,
When the candidate face image is determined to be an appropriate image, the face image authentication unit uses the candidate face image of the query image as a query image, and further compares the query image and the master image to obtain a match. The face image authentication device according to appendix 4, wherein the face image authentication device authenticates when
(Appendix 6)
a binarization processing step of generating a binarized image from the face image based on a predetermined threshold;
a part detection step of detecting a face contour, a right eye region including the right eye, a left eye region including the left eye, a nose region including the nose, and a mouth region including the mouth from the face image;
a first straight line, in the binarized image, having a first start point at any position of the boundary of the right eye region and a first end point at an intersection with the contour of the right face below the first start point; and a detection line determining step of setting a second straight line having a second start point at any position of the boundary of the left eye region and a second end point at an intersection with the contour of the left face below the second start point;
a luminance change detection step of detecting a luminance change of the first straight line and the second straight line, excluding the right eye region, the left eye region, the nose region and the mouth region;
and a determination step of determining that the face image is an image inappropriate for face image authentication when the luminance change is detected for at least one of the first straight line and the second straight line. Appropriate judgment method for face images to be used.
(Appendix 7)
The adequacy determination method according to claim 6, wherein the first straight line is a perpendicular line from the first starting point, and the second straight line is a perpendicular line from the second starting point.
(Appendix 8)
Further, the detection line setting step sets a third straight line connecting the midpoint of the first straight line and the midpoint of the second straight line,
the brightness change detecting step detects brightness changes for the third straight line except for the right eye region, the left eye region, the nose region and the mouth region;
Supplementary note 6 or 7, wherein the determining step determines that the face image is an image inappropriate for face image authentication even when the luminance change is detected for the third straight line. Adequacy judgment method.
(Appendix 9)
With respect to the candidate face image of the master image, if the luminance change is detected, the candidate face image is determined to be an image inappropriate for the face image authentication, and if the luminance change is not detected, the candidate face image is determined. a suitability determination step of determining that the face image is an image suitable for face image authentication;
a storage step of storing the candidate face image of the master image as a master image when the candidate face image is determined to be an appropriate image;
A face image authentication step of matching the query image and the master image and authenticating if they match,
A face image authentication method, wherein the suitability determination step is the suitability determination method according to any one of Appendices 6 to 8.
(Appendix 10)
The suitability determination step determines that the candidate face image is inappropriate for the face image authentication when the luminance change is detected in the candidate face image of the query image, and the luminance change is detected. If not, determine that the candidate face image is an image suitable for face image authentication,
In the face image authentication step, if the candidate face image is determined to be an appropriate image, the candidate face image of the query image is used as a query image, and the query image and the master image are compared to match each other. The face image authentication method according to appendix 9, wherein authentication is performed when
(Appendix 11)
A program for causing a computer to execute the face image adequacy determination method according to any one of appendices 6 to 8.
(Appendix 12)
A program for causing a computer to execute the face image authentication method according to appendix 9 or 10.
(Appendix 13)
A computer-readable recording medium characterized by recording the program according to supplementary note 11 or 12.

According to the present invention, for example, regardless of whether the photograph is blurred or not, it is possible to easily determine a face image that is inappropriate for image authentication, so that it is most suitable for face image authentication.

1 suitability determination device 10 face image acquisition unit 11 binarization processing unit 12 part detection unit 13 detection line setting unit 14 luminance change detection unit 15 determination unit 101 CPU
102 memory 103 interface 104 input device 105 display 106 communication device 107 storage device 108 program 20 binarized image 30 face image 121 (before binarization processing) face contour 122a right eye region 122b left eye region 123 nose region 124 mouth region 131 First straight line 132 Second straight line 133 Third straight line

Claims (10)

a face image acquisition unit that acquires a face image;
a binarization processing unit that generates a binarized image from the face image based on a predetermined threshold;
a part detection unit for detecting a face contour, a right eye area including the right eye, a left eye area including the left eye, a nose area including the nose, and a mouth area including the mouth from the face image;
a first straight line, in the binarized image, having a first start point at any position of the boundary of the right eye region and a first end point at an intersection with the contour of the right face below the first start point; and a detection line setting unit for setting a second straight line having a second start point at any position of the boundary of the left eye region and a second end point at an intersection with the contour of the left face below the second start point;
a brightness change detection unit configured to detect brightness changes of the first straight line and the second straight line except for the right eye area, the left eye area, the nose area, and the mouth area;
and a determination unit that determines that the face image is an image inappropriate for face image authentication when the luminance change is detected for at least one of the first straight line and the second straight line. Appropriateness determination device for facial images. The first straight line is a perpendicular line from the first starting point, the second straight line is a perpendicular line from the second starting point, and the perpendicular line is perpendicular to a line connecting the right eye and the left eye. 2. The adequacy determination device according to claim 1 , wherein: Further, the detection line setting unit sets a third straight line connecting the midpoint of the first straight line and the midpoint of the second straight line,
the brightness change detection unit detects brightness changes for the third straight line except for the right eye region, the left eye region, the nose region and the mouth region;
3. The determination unit determines that the face image is inappropriate for face image authentication even when the luminance change is detected for the third straight line. Appropriateness judgment device. With respect to the candidate face image of the master image, if the luminance change is detected, the candidate face image is determined to be an image inappropriate for the face image authentication, and if the luminance change is not detected, the candidate face image is determined. a suitability determination unit that determines that the face image is suitable for face image authentication;
a storage unit that stores the candidate face image of the master image as a master image when the candidate face image is determined to be an appropriate image;
a face image authentication unit that compares the query image and the master image and authenticates when they match,
A face image authentication device, wherein the suitability determination unit is the suitability determination device according to any one of claims 1 to 3. The suitability determination unit determines that the candidate face image is inappropriate for face image authentication when the luminance change is detected in the candidate face image of the query image, and determines that the luminance change is detected. If not, determine that the candidate face image is an image suitable for face image authentication,
When the candidate face image is determined to be an appropriate image, the face image authentication unit uses the candidate face image of the query image as a query image, and further compares the query image and the master image to obtain a match. 5. The face image authentication device according to claim 4, wherein authentication is performed when the face image is authenticated. a binarization processing step of generating a binarized image from the face image based on a predetermined threshold;
a part detection step of detecting a face contour, a right eye region including the right eye, a left eye region including the left eye, a nose region including the nose, and a mouth region including the mouth from the face image;
a first straight line, in the binarized image, having a first start point at any position of the boundary of the right eye region and a first end point at an intersection with the contour of the right face below the first start point; and a detection line determining step of setting a second straight line having a second start point at any position of the boundary of the left eye region and a second end point at an intersection with the contour of the left face below the second start point;
a luminance change detection step of detecting a luminance change of the first straight line and the second straight line, excluding the right eye region, the left eye region, the nose region and the mouth region;
and a determination step of determining that the face image is an image inappropriate for face image authentication when the luminance change is detected for at least one of the first straight line and the second straight line. Appropriate judgment method for face images to be used. With respect to the candidate face image of the master image, if the luminance change is detected, the candidate face image is determined to be an image inappropriate for the face image authentication, and if the luminance change is not detected, the candidate face image is determined. a suitability determination step of determining that the face image is an image suitable for face image authentication;
a storage step of storing the candidate face image of the master image as a master image when the candidate face image is determined to be an appropriate image;
A face image authentication step of matching the query image and the master image and authenticating if they match,
A face image authentication method, wherein the suitability determination step is the suitability determination method according to claim 6 . 7. A program for causing a computer to execute the face image adequacy determination method according to claim 6. A program for causing a computer to execute the face image authentication method according to claim 7. 10. A computer-readable recording medium recording the program according to claim 8 or 9.

Images