JP6994900B2 - Entrance / exit management system - Google Patents

Description

The present invention relates to an entrance / exit management system that manages entry / exit of workers at an entrance / exit gate at a construction site or the like.

When managing the entrance / exit gate of a construction site where many workers gather, it is conceivable to configure the entrance / exit management system by personal authentication using the face image of each worker.

In the entrance / exit management system that confirms that the target person is the person by the face image, the face photograph of the target person is registered as master data in advance. The degree of matching between the face image of the master data and the face image taken at the entrance gate is called the collation rate. The collation rate is calculated when face recognition is performed, and if the collation rate is equal to or higher than the threshold value, admission permission authentication (authentication success) is performed. If the collation rate does not reach the threshold value, admission is not authenticated (authentication fails).

In such an entrance / exit management system, if there are many cases where authentication is not successful at the time of entry, individual measures may be taken such as lowering the threshold value of the master data of the face image to widen the range of authentication. (For example, refer to Patent Document 1 by the present inventors (Japanese Patent Laid-Open No. 2017-107406)).
JP-A-2017-107406

At construction sites and the like, workers often wear helmets, hats, and masks at the time of admission, and there is a problem that the collation rate tends to decrease. In addition, since the construction site is outdoors, there is also the problem that the collation rate decreases due to the fact that face image data with different light exposure and different exposure is acquired each time during shooting. .. At present, no countermeasures have been taken in consideration of such problems.

The present invention solves the above-mentioned problems, and the entrance / exit management system according to the present invention is an entrance / exit management system that manages the entrance / exit of workers at an entrance / exit gate such as a construction site. A worker face image database registered by associating at least the worker ID with the worker's face image data, and a face image data acquisition unit installed at the entrance / exit gate to acquire the worker's face image data at the time of entry. A collation rate calculation unit that calculates a collation rate between the face image data registered in the worker face image database and the face image data acquired by the face image data acquisition unit, and the face image data acquisition unit. Of the admission recording unit registered by associating the face image data acquired by the unit with the collation rate and the face image data recorded by the admission recording unit, the collation rate is less than the first predetermined value. Moreover, it has a complementary unit that complements the worker face image database with face image data that is lower than the first predetermined value and is equal to or higher than the second predetermined value, and the complementary unit has a first collation rate. The face image data that is less than the predetermined value and is equal to or higher than the second predetermined value lower than the first predetermined value is defined as the face image data of the first group, and the collation rate is less than the second predetermined value. The face image data is used as the face image data of the second group, and the collation rate for each face image data (excluding own data) belonging to the first group and the second group is calculated for all the face image data belonging to the first group. The face image data belonging to the first group having the highest value of (the number of face image data whose collation rate is equal to or higher than the first predetermined value) / (the number of all face image data for which the collation rate is calculated) is complemented. It is characterized in that it is selected as face image data for .

Further, the entrance / exit management system according to the present invention is characterized in that the entrance permission is determined when the collation rate is equal to or higher than the first predetermined value.

In the entrance / exit management system according to the present invention, among the face image data recorded by the entrance recording unit, the collation rate is less than the first predetermined value, and the second is lower than the first predetermined value. Since it has a complementary unit that complements the worker's face image database with face image data that is equal to or more than a predetermined value, according to the entrance / exit management system according to the present invention, when a worker enters, a helmet or a helmet is provided. Even if a hat or mask is worn or facial image data with different exposures is acquired each time, the complemented worker facial image database can prevent the collation rate from decreasing.

It is a figure which shows the outline of the entrance / exit management system 1 which concerns on embodiment of this invention. It is a figure which shows the flowchart of the new registration process of the entrance / exit management system 1 which concerns on embodiment of this invention. It is a figure which shows the display example of the touch panel part 24 in the tablet type terminal 20 with an image pickup function. It is a figure which shows the display example of the touch panel part 24 in the tablet type terminal 20 with an image pickup function. It is a figure which shows an example of the data structure of a worker face image database 11. It is a figure which shows the flowchart of the entrance processing in the entrance / exit management system 1 which concerns on embodiment of this invention. It is a figure which shows an example of the data structure of the entry record 13. It is a figure which shows the data example of the entry record 13 of a specific worker. It is a figure which shows the flowchart of the database supplement processing in the entrance / exit management system 1 which concerns on embodiment of this invention. It is a figure which shows an example of the data structure of the complemented worker face image database 11.

Hereinafter, embodiments of the present invention will be described with reference to the drawings. FIG. 1 is a diagram showing an outline of an entrance / exit management system 1 according to an embodiment of the present invention. The entrance / exit management system 1 according to the present invention is managed by being connected to, for example, a tablet terminal 20 with an image pickup function installed at an entrance / exit gate at a work site of construction work and the tablet terminal 20 with an image pickup function via a communication line. It can be configured with a server 10 installed in an office or the like. It should be noted that such a system configuration is not essential for the entrance / exit management system 1 according to the present invention, and other system configurations may be used.

The entrance / exit management system 1 according to the present invention captures its own face image data by operating the tablet terminal 20 with an image pickup function by itself when each worker enters the work site, and the captured face. Based on the collation rate between the image data and the face image data of the master data, it is assumed that entry / exit permission / non-permission is determined and records at the time of entry are kept. In this embodiment, the process at the time of entry will be described as an example, but the same process can be performed at the time of exit. FIG. 1 shows a worker operating a tablet terminal 20 with an image pickup function by himself / herself.

The tablet terminal 20 with an image pickup function constituting the entrance / exit management system 1 according to the present invention includes a camera unit 22 capable of acquiring an image of an operator, displaying information to the operator, and displaying information to the operator. A general-purpose device having a touch panel unit 24 or the like for receiving an input from is currently widely used.

In this embodiment, the tablet terminal 20 with an image pickup function is used as the information processing terminal used in the entrance / exit management system 1, but an information processing terminal such as a personal computer with a camera, a smartphone, or a mobile phone is used. You can also.

The server 10 transmits data to the tablet terminal 20 with an imaging function, receives and collects data transferred from the tablet terminal 20 with an imaging function, collects and analyzes the data, and records or records the information. Can be displayed. As the server 10, a general-purpose server having a data processing function, a communication function, a data storage function, and the like can be used.

At least two databases, a worker face image database 11 (master data) and an entrance record 13, are recorded in the storage device of the server 10.

The worker face image database 11 is a database in which the face image data of the user and the data related to the user are associated with each other. In the entrance / exit management system 1 according to the present invention, it is assumed that each worker is registered in the worker face image database 11 before entering the work site for the first time.

The process of newly registering each worker by the entrance / exit management system 1 according to the present invention will be described. FIG. 2 is a diagram showing a flowchart of a new registration process of the entrance / exit management system 1 according to the embodiment of the present invention.

FIG. 3 is a diagram showing a display example of the touch panel unit 24 in the tablet terminal 20 with an image pickup function, and is a diagram showing an example of an initial screen of the entrance / exit management system 1. When the worker selects "new registration" on such an initial screen, the new registration process is started in step S100 of FIG.

In step S101 following step S100, the camera unit 22 acquires the facial image of the worker. FIG. 4 is a diagram showing a display example of the touch panel unit 24 in the tablet terminal 20 with an image pickup function. In the tablet terminal 20 with an image pickup function, the face image of the worker acquired by the camera unit 22 is displayed on the touch panel unit 24 as shown in FIG. 5, and the image that makes it easy to recognize the facial features of the worker is automatically displayed. get. As the algorithm for acquiring such an image, a conventionally known algorithm can be used.

In step S102, a unique ID is issued. As the algorithm for issuing such an ID, a conventionally known algorithm can be used.

Subsequently, in step S103, the name of the worker is registered. Such a name registration process can be performed by a worker using the input unit 16 of the tablet terminal 20 with an image pickup function.

By going through the series of steps as described above, it is possible to construct the worker master database 30. FIG. 5 is a diagram showing an example of the data structure of the worker face image database 11. As shown in FIG. 5, the data in the worker face image database 11 includes unique "worker ID" data, "worker name" data, "face image data" registered by the worker for the first time, and a face image. The "registration date and time" data in which the data was registered is recorded.

In the entrance / exit management system 1 according to the present invention, the face image data recorded in the worker face image database 11 and the face image data acquired by the tablet terminal 20 with an image pickup function when the worker enters / exits the entrance / exit. Face recognition is performed based on. As the software for face recognition processing, for example, "NeoFace" (registered trademark) manufactured by NEC Corporation can be used.

Next, the processing at the time of admission after the worker face image database 11 (master data) as described above is created will be described. FIG. 6 is a diagram showing a flowchart of entrance / exit processing in the entrance / exit management system 1 according to the embodiment of the present invention.

When the admission process is started in step S200, the face image data for collation at the time of admission is subsequently imaged in step S201. As shown in FIG. 1, it is assumed that the worker himself / herself acquires such face image data by the tablet terminal 20 with an image pickup function.

In step S202, the face image data registered in advance is acquired from the worker face image database 11, and the collation rate between the face image data registered in advance and the face image data acquired in step S201 is calculated. do.

In step S203, it is determined whether or not the collation rate is N% (70%) or more. N is also defined as a first predetermined value. In the following description, N = 70 is assumed, but N can be set to an appropriate value each time depending on the case.

If the determination in step S203 is YES, the process proceeds to step S204 and admission is permitted. On the other hand, if the determination in step S203 is NO, the process proceeds to step S205 and admission is denied. In the case of the admission disapproval determination in S205, it is preferable to provide a process for urging the worker who is going to enter to take a picture again.

In step S206, the face image data acquired in step S201 is stored in the storage device (not shown) of the server 10 as a history of the face image data acquired in step S201 together with the collation rate and the like. FIG. 7 is a diagram showing an example of the data structure of the entry record 13.

As shown in FIG. 7, in the entrance record 13, for example, the data related to the "worker ID" of the worker, the "worker name" data, the data related to the "date and time", and the "date and time" acquired by the camera unit 22. "Face image" data, data related to the calculated "collation rate", and data related to whether admission is permitted or not are stored in association with each other.

In step S207, the process at the time of admission is completed.

By repeating the above-mentioned process at the time of admission, various collation rate data, face image data at that time, and the like are accumulated in the record 13 at the time of admission.

At construction sites and the like, workers often wear helmets, hats, and masks at the time of admission, and there is a problem that the collation rate tends to decrease. In addition, since the construction site is outdoors, there is also the problem that the collation rate decreases due to the fact that face image data with different light exposure and different exposure is acquired each time during shooting. ..

Therefore, in the entrance / exit management system 1 according to the present invention, among the facial image data stored in the entrance record 13, the collation rate is less than N% (70%) and T% (60%) or more. Among the face image data, the optimum one is added to the worker face image database 11 to complement the data. Hereinafter, N = 70 and T = 60 are set, but N and T can be set to appropriate values in each case. Further, T is defined as a second predetermined value. The second predetermined value (T) is a value lower than the first predetermined value (N).

FIG. 8 is a diagram showing an example of data of the entrance record 13 of a specific worker accumulated in the entrance record 13 by repeating the entrance process. As can be seen in FIG. 8, the face image data whose admission is permitted when the collation rate is N% (70%) or more and the face image data whose admission is denied when the collation rate is less than N% (70%). It is mixed.

As a candidate for face image data to be added to and complemented by the worker face image database 11, face image data having a collation rate of less than N% (70%) and T% (60%) or more is used. ..

Face image data having a collation rate of less than N% (70%) and T% (60%) or more is defined as face image data belonging to the first group. On the other hand, the face image data having a lower collation rate than this, that is, the face image data having a collation rate of less than T% (60%) is defined as the face image data belonging to the second group.

According to the example of FIG. 8, the face image data belonging to the first group having a collation rate of less than N% (70%) and T% (60%) or more are "face image data 2" and "face image data". 4 ”and“ face image data 5 ”. Further, the face image data belonging to the second group having a collation rate of less than T% (60%) are "face image data 7", "face image data 9", "face image data 11", and "face image data 13". , "Face image data 14".

Next, the face image data belonging to the first group having a collation rate of less than N% (70%) and T% (60%) or more is added to the worker face image database 11 to complement the face. The algorithm for selecting image data will be described.

FIG. 9 is a diagram showing a flowchart of database complement processing in the entrance / exit management system 1 according to the embodiment of the present invention.

When the database completion process is started in step S300, the process proceeds to step S301. In step S301, the face image data group (group 1) having a collation rate of less than N% (70%) and T% (60%) or more with respect to the face image data recorded in the entrance record 13. To get. In the example of FIG. 8, in this step, "face image data 2", "face image data 4", and "face image data 5" are acquired.

Subsequent step S302 acquires a face image data group (second group) having a collation rate of less than T% (60%) with respect to the face image data recorded in the entrance record 13. In the example of FIG. 8, in this step, "face image data 7", "face image data 9", "face image data 11", "face image data 13", and "face image data 14" are acquired. Become.

In step S303, one face image data is acquired from the acquired first group. Here, as an example, it is assumed that "face image data 2" has been acquired.

Subsequent step S304 calculates a collation rate for each face image data (excluding own data) belonging to the first group and the second group with respect to the face image data acquired in step S303. The specific processing in this step will be described with reference to FIG. Assuming that the face image data acquired in step S303 is "face image data 2", in step S304, the collation rate between "face image data 2" and "face image data 4", "face image data 2" ", The collation rate between" face image data 5 ", the collation rate between" face image data 2 "and" face image data 7 ", ... The calculation process is executed.

In step S305, R is calculated by [the number of face image data having a collation rate of N% (70%) or more] / [the number of all face image data calculated].

In step S306, it is determined whether or not all the facial image data of the first group have been calculated. If the determination of the step is YES, the process proceeds to step S308, and if the determination is NO, the process proceeds to step S307.

For example, if only "face image data 2" has been calculated in step S306, the process proceeds to step S307, and the next first group of face image data "face image data 4" is acquired, and the step is taken. The calculation of S304 and S305 is repeated.

In step S308, the worker face image database 11 is complemented with the face image data of the first group having the highest R, and in step S309, the database complement processing is completed.

FIG. 10 is a diagram showing an example of the data structure of the worker face image database 11 complemented by the database complement processing as described above. As the data structure shown in FIG. 10, for example, complemented worker face image data and data such as the date and time when the worker face image data is complemented are added to the data structure shown in FIG.

When calculating the collation rate in the entrance processing, the collation rate is appropriately used by using both the worker face image data originally registered in the worker face image database 11 and the complemented worker face image data. Can be increased.

As described above, in the entrance / exit management system 1 according to the present invention, among the face image data recorded by the entrance recording unit, the collation rate is less than the first predetermined value (N), and the first Since it has a complement portion for complementing the worker face image database 11 with face image data having a second predetermined value (T) lower than the predetermined value (N) of 1 and having a face image data equal to or higher than the second predetermined value (T), such an entry according to the present invention. According to the exit management system 1, even if a worker wears a helmet, a hat, or a mask at the time of admission, or if face image data having different exposures is acquired each time, the complemented worker face image database 11 is used. , It is possible to prevent the collation rate from decreasing.

Further, according to the entrance / exit management system 1 according to the present invention, it is possible to prevent a case where the collation rate is poor.

Further, according to the entrance / exit management system 1 according to the present invention, the entrance / exit passage time by face authentication is shortened by increasing the collation rate. As a result, there is no queue.

Further, according to the entrance / exit management system 1 according to the present invention, the authentication rate is improved as the number of complementary face image data registered in the worker face image database 11 increases.

Further, according to the entrance / exit management system 1 according to the present invention, the optimum complementary face image data can always be added according to the result of face authentication, and the authentication rate can be maintained in a high state.

1 ... Entrance / exit management system 10 ... Server 11 ... Worker face image database 13 ... Entrance record 20 ... Tablet-type terminal with imaging function 22 ... Camera unit 24 ... Touch panel Department

Claims (2)

In the entrance / exit management system that manages the entrance / exit of workers at the entrance / exit gates of construction sites, etc.
At least the worker face image database registered by associating the worker ID with the face image data of the worker, and
A face image data acquisition unit installed at the entrance / exit gate to acquire the face image data of the worker at the time of entry, and a face image data acquisition unit.
A collation rate calculation unit that calculates a collation rate between the face image data registered in the worker face image database and the face image data acquired by the face image data acquisition unit, and a collation rate calculation unit.
The entrance recording unit registered by associating the face image data acquired by the face image data acquisition unit with the collation rate, and
Among the face image data recorded by the entrance recording unit, the face image data having a collation rate of less than the first predetermined value and not more than the second predetermined value lower than the first predetermined value is used. It has a complementary part that complements the worker face image database, and has.
The complementary part is
The face image data having a collation rate of less than the first predetermined value and not more than the second predetermined value lower than the first predetermined value is defined as the face image data of the first group.
The face image data whose collation rate is less than the second predetermined value is defined as the face image data of the second group.
For all the face image data belonging to the first group, the collation rate for each face image data (excluding own data) belonging to the first group and the second group was calculated.
The face image data belonging to the first group having the highest value of (the number of face image data whose collation rate is equal to or higher than the first predetermined value) / (the number of all face image data for which the collation rate is calculated) is complemented. An entrance / exit management system characterized by selecting as face image data for . The entrance / exit management system according to claim 1, wherein the entrance permission determination is performed when the collation rate is equal to or higher than the first predetermined value.

Images