JP2023120264A - Collation system - Google Patents

Abstract

To provide a collation system which suppresses a wrong recognition of a target which is going through a gate and also reduces a processing load of an information processor by suppressing unnecessary collation processing.SOLUTION: A collation system 10 includes: acquisition means (an imaging device C1) for acquiring a captured image of a passing front region with respect to a gate; extraction means (a target extraction unit 11) for extracting a facial region of a person to be processed from the captured image; detection means (a target extraction unit 11) for detecting an inter-eye distance of the person from the extracted facial region; collation means (a condition determination unit 12) for starting collation processing of a person registered in advance and the person in the captured image when the inter-eye distance satisfies a predetermined condition; and determination means (a gate control unit 14) for determining whether or not the person can pass through the gate on the basis of a result of the collation processing.SELECTED DRAWING: Figure 2

Description

The present invention relates to a collation system, and more particularly to a collation system that collates objects that are about to pass through a gate.

2. Description of the Related Art As means for restricting or managing persons entering and exiting a specific place such as an office or an event site, a verification system is used to verify whether or not a person trying to pass through is a pre-registered person. In particular, in recent years, with the development of human face recognition technology, a walk-through face recognition system that performs face recognition from a person's face image captured by a camera installed at a gate has been used.

JP-A-2015-1790

Here, in a walk-through face authentication system, there are cases where a plurality of gates are adjacent to each other, and lanes through which persons heading to each gate pass are adjacent to each other. In this case, an image of a person passing through the lane of a certain gate may also include a person heading for the adjacent gate on the adjacent lane. As a result, a person passing through the adjacent lane may be erroneously recognized as a person passing through the own lane, which may hinder verification of a person passing through the own gate ahead of the own lane. For this reason, in situations where a plurality of gates are installed, it is required to suppress misrecognition of a person heading to each gate.

Further, as a technique related to the walk-through face authentication system, there is a technique described in Patent Document 1. Japanese Patent Laid-Open No. 2003-200000 describes a problem that authentication is completed when a person is far from the gate, and the gate is opened before the person reaches the gate. In order to solve such a problem, in Patent Document 1, a photographed face image of a person is compared with a registered face image, and based on the size of the matched person on the input image, the gate of the person is selected. approach is detected.

However, in Japanese Patent Application Laid-Open No. 2002-200011, matching processing is performed on all persons appearing in an input image, and then the approach of a person who has been successfully matched to a gate is detected. Therefore, there arises a problem that the processing load of the collation processing is constantly applied to the information processing apparatus.

In addition, the above-described problem of erroneous recognition of a person heading to each gate and the problem of the processing load of matching processing being constantly imposed on the information processing apparatus are not limited to the case where the target passing through the gate is a person, but any target. It can also occur when performing collation.

Therefore, it is an object of the present invention to solve the above-described problems of erroneously recognizing an object trying to pass through a gate and of constantly imposing a processing load on an information processing apparatus. To provide a collation system capable of

A matching system, which is one embodiment of the present invention,
an imaging means for acquiring a photographed image of a region on the front side of passage with respect to the gate;
a matching means for performing a matching process between a pre-registered target and a target in the photographed image;
Determination means for determining whether or not the object can pass through the gate based on the result of the collation process;
with
The collation means starts the collation process based on a condition set for each area of the captured image regarding the object located in the area.
take the configuration.

Further, an information processing device according to one aspect of the present invention includes:
a matching means for performing a matching process between an object in the photographed image of the area before passing the gate and a pre-registered object;
Determination means for determining whether or not the object can pass through the gate based on the result of the collation process;
with
The collation means starts the collation process based on a condition set for each area of the captured image regarding the object located in the area.
take the configuration.

Further, a program that is one embodiment of the present invention is
information processing equipment,
a matching means for performing a matching process between an object in the photographed image of the area before passing the gate and a pre-registered object;
Determination means for determining whether or not the object can pass through the gate based on the result of the collation process;
It is a program to realize
The collation means starts the collation process based on a condition set for each area of the captured image regarding the object located in the area.
take the configuration.

Further, a matching method, which is one embodiment of the present invention,
The information processing device
Performing matching processing between an object in the photographed image of the area before passing the gate and a pre-registered object,
A method for determining whether the object can pass through the gate based on the result of the collation process,
The information processing device starts the matching process based on a condition set for each area of the captured image regarding the object located in the area.
take the configuration.

According to the present invention configured as described above, it is possible to suppress erroneous recognition of an object trying to pass through a gate, and suppress unnecessary collation processing to reduce the processing load of the information processing apparatus. can be done.

It is a figure which shows the usage condition of the face authentication system in Embodiment 1 of this invention. 1 is a block diagram showing the configuration of a face authentication system according to Embodiment 1 of the present invention; FIG. FIG. 2 is a diagram showing an imaging situation by the face authentication system disclosed in FIG. 1; FIG. 4 is a diagram showing a photographed image photographed in the imaging situation of FIG. 3 by the face authentication system disclosed in FIG. 1; FIG. FIG. 2 is a diagram showing an imaging situation by the face authentication system disclosed in FIG. 1; FIG. 6 is a diagram showing a photographed image photographed in the imaging situation of FIG. 5 by the face authentication system disclosed in FIG. 1; FIG. 2 is a flowchart showing processing operations by the face authentication system disclosed in FIG. 1; It is a figure which shows the imaging condition by the face authentication system in Embodiment 2 of this invention. FIG. 9 is a diagram showing a photographed image photographed in the imaging situation of FIG. 8 by the face authentication system according to Embodiment 2 of the present invention; It is a figure which shows the imaging condition by the face authentication system in Embodiment 2 of this invention. FIG. 11 is a diagram showing a photographed image photographed in the imaging situation of FIG. 10 in the face authentication system according to Embodiment 2 of the present invention; FIG. 11 is a block diagram showing the configuration of a collation system according to Embodiment 3 of the present invention; FIG. 11 is a block diagram showing the configuration of an information processing device according to Embodiment 3 of the present invention;

<Embodiment 1>
A first embodiment of the present invention will be described with reference to FIGS. 1 to 7. FIG. FIG. 1 is a diagram showing usage of a face authentication system. FIG. 2 is a diagram showing the configuration of a face authentication system. 3 to 7 are diagrams for explaining the processing operation of the face authentication system.

[overall structure]
The face authentication system 10 (verification system) in the present invention is a system used to restrict or manage entry and exit of persons (objects) in specific places such as offices and event venues. For example, the imaging device that constitutes the face authentication system 10 is installed near each gate that is opened and closed when a person enters or exits.

In the example shown in FIG. 1, three gates G1, G2, and G3 are arranged in parallel and adjacent to each other. is configured to Therefore, the area on the right side of FIG. 1 with respect to each of the gates G1, G2, and G3 is an area in front of the person passing through the gate (passage front side area). In the front area of each gate G1, G2, G3, lanes R1, R2, R3 along which people who are going to pass through the gates G1, G2, G3 pass side by side are provided. They are located in parallel corresponding to G3. The lanes R1, R2, and R3 may or may not be partitioned by some member.

In the situation shown in FIG. 1, the imaging devices C1, C2, and C3 constituting the face authentication system 10 in the present embodiment are located near the corresponding gates G1, G2, and G3, and the person heading toward the gates G1, G2, and G3 It is located on the right side when viewed from the However, the installation position of the imaging device is not limited to the position shown in FIG. 1, and may be installed in any position such as the left side facing the gate or above the gate. The face authentication system 10 also includes a display device near each imaging device C1, C2, C3.

Then, for example, at the gate G1, the face authentication system 10 captures an image of a person heading toward the gate G1 with the equipped imaging device C1. Then, from the face image of the person appearing in the photographed image, it is collated whether or not the person is a pre-registered person, and if the collation is successful, the gate G1 is opened so that the person can pass through. The configuration of the face authentication system 10 will be described in detail below. Although the face authentication system 10 installed corresponding to the gate G1 will be described below, the face authentication systems installed corresponding to the other gates G2 and G3 have the same configuration.

[Configuration of face authentication system]
The face authentication system 10 in this embodiment is an information processing device having an arithmetic device and a storage device that are integrally configured with an imaging device C1 (camera) and a display device D1 (display). Conversely, the imaging device C1 is equipped with an information processing device having an arithmetic device and a storage device for performing face authentication processing, and a display device D1. However, the face authentication system 10 is not necessarily limited to being configured integrally with the imaging device C1 and the display device D1. For example, the imaging device C1, the display device D1, and the information processing device that processes the captured image may be configured as separate devices and installed at separate locations.

Specifically, as shown in FIG. 2, the face authentication system 10 includes an imaging device C1 and a display device D1. , a matching unit 13 , and a gate control unit 14 . The face authentication system 10 also includes a matching condition storage unit 15 and a matching data storage unit 16 built in a storage device.

The imaging device C1 (imaging means) is provided with a camera and a camera control unit for acquiring a photographed image of the area in front of the gate G1, that is, the area in front of the gate G1 of the corresponding lane R1 at a predetermined frame rate. be. As shown in FIG. 3, for example, the imaging device C1 has an imaging area defined by lines C1a. At this time, the image is captured so that the person P10 passing through the lane R1 (own lane) corresponding to the gate G1 in which the image capturing device C1 is installed is positioned in the center area in the horizontal direction between the reference C1b lines in the image capturing area. A shooting area of the device C1 is set. Note that the captured image is set so as to be generally in focus within a preset distance range from the imaging device C1 in the perspective direction, for example, the distance between the L1 line and the L2 line shown in FIG.

By setting the photographing area of the imaging device C1 as described above, the end areas in the horizontal direction with respect to the central area, which is the area sandwiched between the reference C1a line and the reference C1b line, in the photographing area. In the situation shown in FIGS. 3 and 4, a person P20 passing through the lane R2 corresponding to the adjacent gate G2 may appear. In addition, in the situations shown in FIGS. 5 and 6, the person P12 positioned in the own lane R1 corresponding to the own gate G1 where the imaging device C1 is installed may appear in the end region. Even if such a situation occurs, the face authentication system 10 of the present invention appropriately recognizes the person P12 who is about to pass through the own gate G1 and determines whether or not the person P12 can pass through the gate, as described below. It is configured to be able to determine and further reduce the processing load.

The target extracting unit 11 (target information detecting means) acquires a photographed image from the imaging device C1, and determines determination information (target information) used for determining whether or not to start the matching process, for example, as follows. to detect. First, the face area of the person to be processed is extracted from the photographed image. Extraction of a person's face area is performed, for example, by judging from the position, color, etc. with respect to the whole image of a moving person. Then, the target extracting unit 11 detects the features of the person as determination information from the face area. In this embodiment, the feature as determination information is the distance between the eyes of a person (inter-eye distance), which is detected as symbols d10, d20, and d12 as shown in FIGS. However, the inter-eye distance detection method is not limited to the method described above, and any method may be used for detection. Also, as will be described later, the determination information may be other information representing characteristics of a person, and may be detected by any method.

The condition determination unit 12 (collation means) determines whether the inter-eye distances d10, d20, and d12, which are the features of the person as the determination information detected by the object extraction unit 11, satisfy the condition for starting the collation process. judge. Here, in the present embodiment, different values are set for the conditions for starting the matching process for each region in the captured image where the person is located, that is, different values are set for the central region and the end regions in the captured image. Such values are pre-stored in the matching condition storage unit 15 . For example, as the value Tc of the condition corresponding to the central region of the captured image, a value is set such that when the person is positioned in the lane R1 of the own gate G1, the distance between the eyes of the person is always exceeded. On the other hand, as the value Te of the condition corresponding to the edge region of the captured image, a value larger than the value Tc of the condition set for the central region is set (Te>Tc). As an example, as the value Te of the condition corresponding to the edge area, as shown in FIG. 5 and FIG. A value that is small and larger than the inter-eye distance d20 of the person P20 positioned in the adjacent lane R2 as shown in FIGS. 3 and 4 is set.

In the above situation, the condition determination unit 12 determines that the inter-eye distances d10, d20, and d12, which are features of a person as determination information, are the condition values (Tc or Te) is exceeded. For example, in the case of FIGS. 3 and 4, for the person P10 positioned in the central area of the captured image, it is determined whether the inter-eye distance d10 exceeds the conditional value Tc set for the central area. Also, for the person P20 located in the edge region of the captured image, it is determined whether or not the inter-eye distance d20 exceeds the conditional value Te set for the edge region. Similarly, in the case of FIGS. 5 and 6, for the person P10 located in the central area of the captured image, it is determined whether the inter-eye distance d10 exceeds the conditional value Tc set for the central area. Also, for the person P12 located in the edge region of the captured image, it is determined whether or not the inter-eye distance d12 exceeds the conditional value Te set for the edge region. In the examples of FIGS. 3 and 4, the person P10 in the center area satisfies the condition, and the person P20 in the end area does not satisfy the condition. In the examples of FIGS. 5 and 6, the person P10 in the central area satisfies the condition, and the person P12 in the edge area also satisfies the condition.

The collation unit 13 (collation means) performs the collation unit 13 when the condition determination unit 12 determines that the distance between the eyes of the person exceeds the value of the condition set in the area where the person is located, that is, the condition is satisfied. Then, the matching process for such person is started. In other words, the collation unit 13 does not perform collation processing for persons who did not satisfy the condition determined by the condition determination unit 12 . Here, the matching process is performed, for example, as follows. First, a feature amount necessary for matching is generated from the face area of a person to be matched. Then, a matching score, which is the degree of similarity between the generated feature amount and the person's feature amount registered in advance in the matching data storage unit 16, is calculated, and it is determined whether or not the matching score is higher than the threshold value. If the collation score is higher than the threshold, it is determined that the collation is successful, and that the person who is about to pass through the gate G1 is a pre-registered person. At this time, the feature amount of the person detected for matching may be the feature amount used in the existing face matching technology, or may be the feature amount calculated by any method. Moreover, any collation method may be used.

The gate control unit 14 (determining means) first determines whether or not a person can pass through the gate G1 based on the result of collation by the collating unit 13 . Specifically, it is determined that a person for whom collation by the collating unit 13 is successful is passable. In addition, the gate control unit 14 has a function of displaying the collation result, that is, the success or failure of the collation on the display device D1. Further, the gate control unit 14 also has a gate control function of opening and closing the gate G1, and controls the gate G1 to be opened for a person determined to be able to pass through.

The display device D1 is arranged with its display surface facing the area in front of the passage of the gate G1 so that a person who is about to pass through the gate G1 can see it. However, the display device D1 may not necessarily be provided.

[motion]
Next, the operation of the face authentication system 10 described above will be described with reference to the flowchart of FIG. Here, the operation of the face authentication system 10 corresponding to the gate G1 will be described, and an example will be described in which the imaging conditions of the imaging device C1 are the conditions shown in FIGS. 3 to 6. FIG.

The imaging device C1 corresponding to the gate G1 continues to capture the image of the area in front of the passage of the gate G1. Then, the face authentication system 10 always performs the following processing on the captured image.

First, the object extracting unit 11 extracts a person (object) to be processed from within the photographed image (step S1). Then, from the extracted person, the feature of the person, here, the distance between the eyes of the person, is detected as determination information necessary for determining the conditions for starting the matching process (step S2). For example, they are detected as symbols d10 and d20 in FIG. 4 and symbols d10 and d12 in FIG.

Subsequently, the condition determination unit 12 determines whether or not the inter-eye distance, which is the detected determination information, satisfies the conditions for starting the matching process. At this time, since the value of the condition is set for each region where the person is located, the value of the condition corresponding to the position of the person is read from the collation condition storage unit 15, and the detected distance between the eyes determines the value of the condition. It is determined whether or not it exceeds (step S3). In the examples of FIGS. 3 and 4, it is determined that the inter-eye distance d10 of the person P10 positioned in the central area of the captured image exceeds the conditional value Tc set for the central area. Then, it is determined that the eye-to-eye distance d20 of the person P20 positioned in the edge region of the captured image does not exceed the conditional value Te set for the edge region, which is a value greater than the value Tc. Therefore, it is determined that only the person P10 satisfies the condition for starting the matching process (Yes in step S3). Further, in the examples of FIGS. 5 and 6, it is determined that the inter-eye distance d10 of the person P10 positioned in the central area of the captured image exceeds the conditional value Tc set for the central area. Then, it is determined that the distance d12 between the eyes of the person P12 positioned in the edge region of the captured image exceeds the value Te of the condition set for the edge region. Therefore, in the examples of FIGS. 5 and 6, it is determined that the person P10 and the person P12 satisfy the conditions for starting the matching process (Yes in step S3).

Subsequently, the collation unit 13 starts collation processing for a person whose eye-to-eye distance has been determined by the condition determination unit 12 to exceed the value of the condition set for each region (step S4). In the example of FIGS. 3 and 4, the matching process for only the person P10 is started, and in the example of FIGS. 5 and 6, the matching process for the person P10 and the person P12 is started. In the matching process, first, the feature amount necessary for matching is detected from the face area of the person to be matched, and the feature amount of the person registered in advance in the matching data storage unit 16 is matched to obtain a matching score. do. Then, it is determined whether or not the collation score is equal to or greater than the threshold. In the examples of FIGS. 5 and 6, the person P10 and the person P12 are to be matched. It is also possible to determine whether or not the passage is permitted and to open and close the gate.

As a result of the collation processing by the collation unit 13, when the collation of the person trying to pass is successful (Yes in step S5), the gate control unit 14 (determining means) permits the person to pass through the gate G1, and opens the gate G1. control (step S6). At this time, the gate control device 14 displays on the display device D1 whether or not the passage is permitted.

As described above, according to the face authentication system 10 of the present embodiment, it is possible to suppress the matching process for the person P20 who is about to pass through the adjacent gate G2, and appropriately perform the matching process for the people P10 and P12 who are about to pass through the own gate G1. can do. For example, in the examples of FIGS. 3 and 4, the matching process is performed for the person P10 who is about to pass through the own gate G1, but the matching process is not performed for the person P20 who is about to pass through the adjacent gate G2. In addition, in the examples of FIGS. 5 and 6, even the person P12 appearing in the edge area of the captured image is about to pass through the own gate G1, so the matching process is appropriately performed. As a result, erroneous recognition of a person trying to pass through another gate can be suppressed, and unnecessary matching processing can be suppressed, thereby reducing the processing load of the information processing apparatus.

In the above description, the case where the area in the captured image is horizontally divided into the center area and the end area is exemplified, but the area may be divided into other areas. Also in this case, conditions for starting the matching process are set for each area.

Further, in the above description, the case where the determination information, which is the feature of the person to be compared with the value of the condition for starting the matching process, is the inter-eye distance was exemplified. It may be a distance or a value representing the size of another part of the person. Accordingly, the value of the condition for starting the matching process is also set for each region in the captured image.

Also, in the above description, the case where the object trying to pass through the gate G1 is a person is exemplified, but the object is not limited to a person and may be any object. For example, it may be an object such as luggage. Accordingly, the determination information representing the feature to be compared with the value of the condition for starting the matching process may be information representing any feature that can be detected from the object. Moreover, any feature amount that can be detected from an object may be used when performing the matching process.

<Embodiment 2>
Next, a second embodiment of the invention will be described with reference to FIGS. 8 to 11. FIG. 8 to 11 are diagrams for explaining the processing operation of the face authentication system.

The face authentication system 10 of this embodiment has the same configuration as that of the above-described first embodiment, but the criteria for determining whether or not the conditions for starting the matching process are satisfied are different. That is, in the present embodiment, the orientation of the person with respect to the gate G1 is detected from the captured image as determination information, and when this determination information satisfies the conditions set for each area of the captured image, the matching process is started. . Configurations different from those of the first embodiment will be mainly described in detail below.

The target extraction unit 11 (target information detection means) in the present embodiment first extracts the face region of the person to be processed from within the captured image. Then, the target extraction unit 11 detects the orientation of the person's face with respect to the own gate G1 as the person's feature, which is determination information used for determining whether or not to start the matching process. Specifically, as the determination information, the angle at which the front of the face faces with respect to the optical axis A1 of the imaging device C1 is detected. For example, parts such as the eyes, nose, and mouth that make up a human face are extracted, the absolute frontal orientation of the face is detected from the positional relationship thereof, and further, from the installation angle of the imaging device C1, the imaging device C1 is detected. It is possible to detect the angle at which the face is facing with respect to Note that the orientation of the face may be detected by any method.

Then, the condition determination unit 12 (collation unit) in the present embodiment performs a collation process in which the angle at which the face faces the imaging device C1, which is the determination information detected by the object extraction unit 11, is set for each region of the captured image. It is determined whether the conditions for starting are satisfied. In this embodiment, different values are set for the condition for starting the matching process between the center region and the end region in the horizontal direction in the photographed image, and such values are stored in the matching condition storage unit 15 in advance. It is

Here, the values of the conditions set corresponding to the center area and the edge area of the captured image will be described. As shown in FIGS. 8 and 9, there is a high possibility that the person P10 located in the central area of the captured image and heading toward the own gate G1 faces the capturing direction of the imaging device C1. Therefore, there is a high possibility that the angle formed by the direction of the optical axis A1 of the imaging device C1 of the person P10 and the direction in which the front face of the person P10 is facing is extremely small. On the other hand, the person P20 located in the end region of the captured image and heading for the adjacent gate G2 is highly likely to face the imaging direction of the imaging device C2 installed at the gate G2. It is highly likely that you are not facing the right direction. Therefore, there is a high possibility that the angle formed by the direction of the optical axis A1 of the imaging device C1 of the person P20 and the direction in which the person's face is facing is extremely large.

In addition to the above, as shown in FIGS. 10 and 11, even if the person P12 is located in the edge area of the captured image, the person P12 may be facing the capturing direction of the imaging device C1 if the person is heading toward the own gate G1. is high. Therefore, the angle between the direction of the optical axis A1 of the imaging device C1 of the person P12 and the direction in which the person's face faces is likely to be smaller than the angle of the person P20 located in the lane R2 of the adjacent gate G2.

Considering the above, first, it is assumed that the value of the condition corresponding to the central region of the captured image is an extremely small value, which is the characteristic of the person P10 heading toward the own gate G1 located in the central region. Therefore, it is set to a small value that includes such angles. Therefore, for example, the value of the condition corresponding to the central region is set to 10°. On the other hand, the value of the condition corresponding to the edge region of the captured image is set larger than the value of the condition corresponding to the central region. This is because the angle, which is the determination information of the person P12 heading toward the own gate G located in the end area, is expected to be larger than that of the person P10 in the center area. However, the value of the condition corresponding to the edge region is set to a value smaller than the angle assumed by the person P20 at the adjacent gate G2. Therefore, for example, the value of the condition corresponding to the edge region is set to 60°. However, the values of the conditions are not limited to the values described above, and are appropriately set according to the situation such as the arrangement of gates.

In the above situation, the condition determination unit 12 determines whether the angle of the face with respect to the imaging device C1, which is the feature of the person as the determination information, is equal to or less than the value of the condition set for the area in the captured image in which the person is positioned. determine whether or not For example, in the case of FIGS. 8 and 9, for the person P10 positioned in the central area of the captured image, it is determined whether the angle is equal to or less than the value of the condition set for the central area, and the end areas of the captured image are determined. As for the person P20 positioned at , it is determined whether the angle is equal to or less than the value of the condition set for the edge area. Similarly, in the case of FIGS. 10 and 11, for the person P10 positioned in the central area of the captured image, it is determined whether the angle is equal to or less than the value of the condition set for the central area. For the person P12 located in the area, it is determined whether the angle is equal to or less than the value of the condition set for the edge area. In the examples of FIGS. 8 and 9, the person P10 in the center area satisfies the condition, and the person P20 in the end area does not satisfy the condition. In the examples of FIGS. 10 and 11, the person P10 in the central area satisfies the condition, and the person P12 in the edge area also satisfies the condition.

Then, when the condition determination unit 12 determines that the angle of the person with respect to the imaging device C1 is equal to or less than the value of the condition set for the area in which the person is located, that is, the condition is satisfied, the collation unit 13 A verification process is started for such a person. In other words, the collation unit 13 does not perform collation processing for persons who did not satisfy the condition determined by the condition determination unit 12 .

As described above, according to the face authentication system 10 of the present embodiment, it is possible to suppress matching processing for a person who is about to pass through the adjacent gate G2, and to appropriately perform matching processing for a person who is about to pass through the own gate G1. . As a result, erroneous recognition of a person trying to pass through another gate can be suppressed, and unnecessary matching processing can be suppressed, thereby suppressing the processing load of the information processing apparatus.

Note that the orientation of the face, which is a characteristic of a person as the determination information described above, is not limited to the orientation of the face with respect to the imaging device C1, and may be a value based on any standard. Also, the value of the condition for starting the matching process is not limited to the values described above, and may be any value.

<Embodiment 3>
Next, a third embodiment of the invention will be described with reference to FIGS. 12 and 13. FIG. FIG. 12 is a block diagram showing the configuration of a collation system according to the third embodiment. FIG. 13 is a block diagram showing the configuration of the information processing apparatus according to the third embodiment. It should be noted that this embodiment shows an outline of the configuration of the face authentication system described in the first and second embodiments.

As shown in FIG. 12, the collation system 100 in this embodiment
Imaging means 110 for acquiring a photographed image of the area on the front side of the passage with respect to the gate;
a matching means 120 for matching a target registered in advance with a target in a photographed image;
Judgment means 130 for judging whether or not the target can pass through the gate based on the result of the collation process;
Prepare.
Then, the collation means 120 starts collation processing based on the conditions set for each area of the photographed image for the object located in the area.
take the configuration.

Further, in this embodiment, the imaging means 110 may be removed from the matching system 100 shown in FIG.
That is, the information processing device 200 in this embodiment
A matching means 210 that performs matching processing between an object in the photographed image of the region on the side before passing the gate and a pre-registered object;
Judgment means 220 for judging whether or not the object can pass through the gate based on the result of the collation process;
Prepare.
Then, the collation means 210 starts collation processing based on the conditions set for each area of the photographed image for the object located in the area.
take the configuration.

The collation means 120 and 210 and the determination means 130 and 220 described above may be constructed by executing a program by an arithmetic unit, or may be constructed by an electronic circuit.

Then, according to the matching system 100 or the information processing device 200 configured as described above,
Performing matching processing between an object in the photographed image of the area before passing the gate and a pre-registered object,
Determining whether the target can pass through the gate based on the result of the matching process;
A matching method is provided.
Then, in the matching method, the matching process is further started based on the conditions set for each area of the captured image regarding the object located in the area.
take the configuration.

According to the matching system 100 or the information processing device 200 described above, for a target located in a predetermined area of a captured image, matching processing is started when the target satisfies the conditions set for the area. Therefore, even if an object about to pass through another gate appears in the photographed image, the start of matching processing for that object can be suppressed by the conditions set for the area in which the object appears. As a result, erroneous recognition of an object trying to pass through the gate can be suppressed, and unnecessary collation processing can be suppressed, thereby reducing the processing load of the information processing apparatus.

<Appendix>
Some or all of the above embodiments may also be described as the following appendices. The outline of the configuration of the collation system, the information processing device, the program, and the collation method according to the present invention will be described below. However, the present invention is not limited to the following configurations.

(Appendix 1)
an imaging means for acquiring a photographed image of a region on the front side of passage with respect to the gate;
a matching means for performing a matching process between a pre-registered target and a target in the photographed image;
Determination means for determining whether or not the object can pass through the gate based on the result of the collation process;
with
The collation means starts the collation process based on a condition set for each area of the captured image regarding the object located in the area.
matching system.

According to the above configuration, the matching system starts the matching process for an object located in a predetermined area of a captured image when the object satisfies the conditions set for the area. Therefore, even if an object about to pass through another gate appears in the photographed image, the start of matching processing for that object can be suppressed by the conditions set for the area in which the object appears. As a result, erroneous recognition of an object trying to pass through the gate can be suppressed, and unnecessary collation processing can be suppressed, thereby reducing the processing load of the information processing apparatus.

(Appendix 2)
The matching system according to Supplementary Note 1,
an object information detecting means for detecting object information representing characteristics of the object in the captured image;
The matching means starts the matching process for the target when the target information of the target in the captured image satisfies a condition set in an area of the captured image in which the target is located.
matching system.

According to the above configuration, the object information representing the characteristics of the object in the photographed image is detected, and when the object information satisfies the condition set for each area of the photographed image, the matching process for the object is started. Therefore, by setting the target information of the target to a condition that does not start the matching process in an area showing the target trying to pass through a different gate, erroneous recognition of the target can be suppressed, and unnecessary matching can be suppressed. can be done.

(Appendix 3)
The matching system according to Supplementary Note 2,
The collation means, when the object information of the object in the photographed image satisfies a condition set for each horizontally divided area of the photographed image in which the object is located, the collation of the object. start processing,
matching system.

In this way, by setting a condition for each region divided in the horizontal direction, erroneous recognition of objects trying to pass through adjacent gates and unnecessary collation processing can be suppressed.

(Appendix 4)
The verification system according to Appendix 3,
The target information detection means detects the size of the target as the target information,
The matching means performs the matching process on the target when the target information of the target in the captured image is larger than a value that is a condition set for each area of the captured image in which the target is located. start and
A value that is a condition for the target information of the target set in an end region with respect to a center region in the horizontal direction of the captured image is larger than a value that is a condition for the target information of the target set in the center region. is set,
matching system.

(Appendix 5)
The verification system according to Appendix 4,
The target information detection means detects, as the target information representing the size of the target, a distance between predetermined parts forming a face of a person who is the target.
matching system.

According to the above configuration, the condition regarding the size of the object appearing in the edge area is set larger than the size of the object moving to the central area. For example, when the target is a person, the condition of the distance between the predetermined parts forming the person's face is set larger for the end regions than for the center region. As a result, when an object about to pass through an adjacent gate appears in the edge area, the size of the object does not satisfy the condition, so erroneous recognition of the object and unnecessary collation processing can be suppressed. On the other hand, when an object about to pass through the corresponding gate appears in the edge region, the size of the object satisfies the condition, and the matching process can be appropriately started.

(Appendix 6)
The verification system according to Appendix 3,
The target information detection means detects an orientation of the target with respect to the gate as the target information,
When the target information of the target in the captured image satisfies a condition set for each region of the captured image in which the target is located, the matching means starts the matching process for the target,
A condition for the object information of the object set in the end region with respect to the central region in the horizontal direction of the photographed image and a condition for the object information of the object set in the central region are set differently. ,
matching system.

(Appendix 7)
The verification system according to Appendix 6,
The object information detection means detects the angle at which the face of the person is facing with respect to the imaging means, based on the position of a predetermined part of the face of the person, as the object information representing the orientation of the object. detect and
The collation means performs the above-mentioned method for the object when the angle, which is the object information, of the object in the photographed image is equal to or less than a value that is a condition set for each area of the photographed image in which the object is located. start the matching process,
A value that is a condition for the target information of the target set in the end region with respect to the center region in the horizontal direction of the captured image is greater than an angle value that is a condition for the target information of the target set in the center region. is also set large,
matching system.

According to the above configuration, different conditions are set for the center area and the edge area regarding the orientation of the object appearing in the edge area. For example, when the object is a person, the condition of the angle at which the person's face is facing with respect to the imaging device is set larger for the end regions than for the central region. As a result, when an object about to pass through the corresponding gate appears in the edge area, the angle of the object satisfies the condition, and the matching process can be appropriately started. On the other hand, when an object about to pass through an adjacent gate appears in the edge area, the angle of the object does not satisfy the conditions, so erroneous recognition of the object and unnecessary collation processing can be suppressed.

(Appendix 8)
a matching means for performing a matching process between an object in the photographed image of the area before passing the gate and a pre-registered object;
Determination means for determining whether or not the object can pass through the gate based on the result of the collation process;
with
The collation means starts the collation process based on a condition set for each area of the captured image regarding the object located in the area.
Information processing equipment.

(Appendix 9)
The information processing device according to appendix 8,
an object information detecting means for detecting object information representing characteristics of the object in the captured image;
The matching means starts the matching process for the target when the target information of the target in the captured image satisfies a condition set in an area of the captured image in which the target is located.
Information processing equipment.

(Appendix 10)
information processing equipment,
a matching means for performing a matching process between an object in the photographed image of the area before passing the gate and a pre-registered object;
Determination means for determining whether or not the object can pass through the gate based on the result of the collation process;
It is a program to realize
The collation means starts the collation process based on a condition set for each area of the captured image regarding the object located in the area.
program.

(Appendix 11)
The program according to Appendix 10,
causing the information processing device to further implement target information detection means for detecting target information representing characteristics of the target in the captured image;
The matching means starts the matching process for the target when the target information of the target in the captured image satisfies a condition set in an area of the captured image in which the target is located.
program.

(Appendix 12)
The information processing device
Performing matching processing between an object in the photographed image of the area before passing the gate and a pre-registered object,
A method for determining whether the object can pass through the gate based on the result of the collation process,
The information processing device starts the matching process based on a condition set for each area of the captured image regarding the object located in the area.
Matching method.

(Appendix 13)
The matching method according to Appendix 12,
The information processing device detects previous target information representing features of the target in the captured image,
The information processing device starts the matching process for the target when the target information of the target in the captured image satisfies a condition set in an area of the captured image in which the target is located.
Matching method.

The program described above is stored in a storage device or recorded in a computer-readable recording medium. For example, the recording medium is a portable medium such as a flexible disk, an optical disk, a magneto-optical disk, and a semiconductor memory.

Although the present invention has been described with reference to the above-described embodiments and the like, the present invention is not limited to the above-described 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.

10 face authentication system 11 object extraction unit 12 condition determination unit 13 matching unit 14 gate control unit 15 matching condition storage unit 16 matching data storage unit 100 matching system 110 imaging means 120 matching means 130 determining means 200 information processing device 210 matching means 220 determination means C1, C2, C3 imaging device D1 display device G1, G2, G3 gate

Claims (10)

a detection means for detecting an object from a captured image;
collation means for performing collation processing between the object in the photographed image and a pre-registered object;
and output means for outputting whether or not the object can pass through the gate based on the result of the collation process,
The matching means does not start matching processing between a pre-registered target and the target in the captured image when the feature information of the target in the captured image does not satisfy a predetermined condition.
matching system. the detection means detects a face area of the target from the captured image;
The matching system according to claim 1. characterized information detection means for detecting the characteristic information of the target from the captured image;
3. The collation system according to claim 1 or 2. The feature information is the distance between the eyes of the target,
A collation system according to any one of claims 1 to 3. The feature information is the orientation of the face of the target,
A collation system according to any one of claims 1 to 3. The collation means compares the object detected by the detection means with a pre-registered object in the photographed image when the feature information of the object in the photographed image does not satisfy a predetermined condition. does not start the matching process with the target,
A collation system according to any one of claims 1 to 5. The output means outputs whether or not the object can pass through the gate based on the result of the matching process and the feature information.
A collation system according to any one of claims 1 to 6. The output means outputs whether or not the object can pass through the gate from the object with the large distance between the eyes.
A matching system according to claim 4. Detect the target from the captured image,
performing matching processing between the target in the captured image and a pre-registered target;
outputting whether or not the object can pass through the gate based on the result of the matching process;
When performing the matching process, if the feature information of the target in the captured image does not satisfy a predetermined condition, the matching process between the pre-registered target and the target in the captured image is not started. ,
Matching method. information processing equipment,
Detect the target from the captured image,
performing matching processing between the target in the captured image and a pre-registered target;
outputting whether or not the object can pass through the gate based on the result of the matching process;
When performing the matching process, if the feature information of the target in the captured image does not satisfy a predetermined condition, the matching process between the pre-registered target and the target in the captured image is not started. ,
A program for executing a process.

Images