A face recognition method and system for an electronic fence

Abstract

The present application discloses a face recognition method for an electronic fence. The moving targets are confirmed through multi-target motion detection technology, and face recognition scanning is performed on the moving targets. Moving targets with face features are screened out to generate a face feature record table. According to the face feature record table, it is analyzed and determined that the personnel who make multiple round trips or stay for too long are suspected objects with intrusion intentions. The present application discloses a face recognition system for an electronic fence. The shooting of the scene is completed through a monitoring camera, the display of the captured image is completed through a display module, and the analysis and confirmation of personnel making multiple round trips and long-term stayers are completed through an analysis and processing module, so as to determine the suspected objects with intrusion intentions, avoiding the manual consumption of manual monitoring and the omission problem in the process of manual monitoring, and improving the security of the application of the electronic fence.

Description

Technical Field

[0001] The present invention relates to the technical field of electronic fence protection, and more particularly to a face recognition method and system for an electronic fence. Background Art

[0002] In some areas that need to be enclosed, such as accident scenes, emergency rescue scenes or production sites, in order to prevent unauthorized personnel from entering, obstacles such as warning lines or fences are usually set up; in order to more conveniently monitor whether there are people entering the enclosed area, a more advanced method is to set up an electronic fence. If an intruder touches the electronic fence, an alarm will be sent to the staff so that they can rush to the scene in time to expel the intruder. However, for a site with a large enclosed area, due to the large peripheral range of the site, there may be a large number of people around the electronic fence, and there may be intruders with the intention of intrusion among them. In order to break through the electronic fence, they will make multiple trips to the scene. In this case, in order to identify the intruders with the intention of intrusion, using a high-alert state monitoring is obviously too labor-consuming. Therefore, there is an urgent need for a method or system that can save labor and effectively prevent intruders. Summary of the Invention

[0003] The purpose of the present invention is to provide a face recognition method and system that can act before the electronic fence to pre-identify intruders with the intention of intrusion and improve the protection performance of the electronic fence.

[0004] Therefore, the present invention discloses a face recognition method for an electronic fence, which is applied with a display module. The method includes:

[0005] Obtain the video information captured by a monitoring camera, determine the moving targets according to multi-target motion detection technology, and mark the moving targets on the display module;

[0006] Perform face recognition scanning on the moving targets. If a face feature is determined to exist, generate a face feature record table according to the face feature, and record the time and frequency of detecting the face feature in the face feature record table;

[0007] If the time length or frequency of detecting the face feature is greater than a preset value, determine the moving target corresponding to the face feature as a suspected intrusion object and mark it on the display module.

[0008] In some embodiments of the present application, in order to be able to judge whether a moving target repeatedly approaches the electronic fence according to the content recorded in the face feature record table, the content of the face record table is disclosed. The content of the face feature record table includes:

[0009] Facial feature values, used to distinguish the moving target;

[0010] Several facial feature discovery times, used to determine the discovery time of the facial features;

[0011] Several facial feature disappearance times, used to determine the disappearance time of the facial features.

[0012] In some embodiments of the present application, a method for determining the time length of monitoring the facial features through the facial feature record table is disclosed. The method for determining the time length of monitoring the facial features includes:

[0013] According to the facial feature discovery time and facial feature disappearance time recorded in the facial feature record table, determine the time length of monitoring the facial features.

[0014] In some embodiments of the present application, a method for determining the frequency of monitoring the facial features through the facial feature record table is disclosed. The method for determining the frequency of monitoring the facial features includes:

[0015] According to the update times of the facial feature discovery time in the facial feature record table recorded within a unit time, determine the frequency of monitoring the facial features.

[0016] In some embodiments of the present application, in order to reduce the workload of face recognition scanning and improve the number of monitored moving targets, the method is improved. The method further includes:

[0017] Preset several levels of warning lines on the periphery of the electronic fence, and analyze and determine the time length for the moving target to cross adjacent two levels of warning lines;

[0018] If the time length for the moving target to cross adjacent two levels of warning lines is less than a preset value, it is determined that the moving target has completed a rapid crossing of the warning line;

[0019] If the number of times the moving target has completed the rapid crossing of the warning line is greater than a preset value, then perform face recognition scanning on the moving target again.

[0020] The present application discloses a face recognition method for an electronic fence, which confirms the moving target through multi-target motion detection technology, performs face recognition scanning on the moving target, generates a facial feature record table for the moving target screened out with facial features, and based on the facial feature record table, analyzes and determines that the person who makes multiple round trips or stays for too long is a suspected object with an intrusion intention, avoiding the manual consumption of manual monitoring and not having the problem of omission in manual monitoring.

[0021] A face recognition system for an electronic fence, the system comprising:

[0022] A monitoring camera for taking pictures of the electronic fence site;

[0023] A display module for displaying the images taken by the monitoring camera;

[0024] An analysis and processing module for analyzing and processing the video information taken by the monitoring camera;

[0025] The analysis method of the analysis and processing module includes determining moving targets according to multi-target motion detection technology, performing face recognition scanning on the moving targets, if face features are determined to exist, generating a face feature record table according to the face features, and recording the time and frequency of detecting the face features in the face feature record table;

[0026] The analysis and processing module calculates the time length and frequency of detecting the face features according to the face feature record table, and compares the time length and frequency of detecting the face features with preset values. If the time length or frequency of detecting the face features is greater than the preset value, it is determined that the moving target corresponding to the face features is a suspected intrusion object, and the display module is driven to mark the suspected intrusion object.

[0027] In some embodiments of the present application, a method capable of analyzing and determining the time length and frequency of detecting the face features is provided. The method for the analysis and processing module to determine the time length and frequency of detecting the face features includes:

[0028] The content of the face feature record table includes face feature values, several face feature discovery times, and several face feature disappearance times;

[0029] Determine the time length of detecting the face features according to the face feature discovery times and face feature disappearance times recorded in the face feature record table;

[0030] Determine the frequency of detecting the face features according to the update times of the face feature discovery times in the face feature record table recorded within a unit time.

[0031] In some embodiments of the present application, in order to reduce the workload of face recognition scanning of the system and increase the number of monitored moving targets, an early warning module is disclosed to realize pre-screening and warning of the moving target points. The system further includes an early warning module, and the early warning module is used to make a preliminary judgment on the moving target to determine whether to perform face recognition scanning on the moving target.

[0032] In some embodiments of the present application, a method for the warning and prevention module to pre-screen and judge a moving target is disclosed. The method for the warning and prevention module to pre-screen and judge the moving target includes:

[0033] A number of levels of warning lines are preset around the electronic fence. The warning and prevention module analyzes and determines the time length for the moving target to cross two adjacent warning lines. If the time length for the moving target to cross two adjacent warning lines is less than a preset value, it is determined that the moving target has completed a rapid crossing of the warning line. If the number of times the moving target has completed a rapid crossing of the warning line is greater than a preset value, the analysis and processing module performs a face recognition scan on the moving target.

[0034] The present application discloses a face recognition system for an electronic fence. The on-site shooting is completed through a monitoring camera, the display of the captured image is completed through a display module, and the analysis and confirmation of multiple round-trip personnel and long-term staying personnel are completed through an analysis and processing module, so as to determine a suspected object with an intrusion intention, avoiding the labor consumption of manual monitoring and the omission problem in the process of manual monitoring, and improving the security of the application of the electronic fence.

[0035] The technical solution of the present invention will be further described in detail below with reference to the drawings and embodiments. Description of the Drawings

[0036] Figure 1 It is a conceptual step diagram of a face recognition system for constructing an electronic fence in an embodiment of the present application;

[0037] Figure 2 It is a method step diagram for warning and screening of a moving target in an embodiment of the present application. Detailed Embodiments

[0038] The technical solution of the present invention will be further described below with reference to the drawings and embodiments.

[0039] Unless otherwise defined, the technical terms or scientific terms used in the present invention shall have the ordinary meanings as understood by those of ordinary skill in the field to which the present invention pertains. The terms "first", "second" and similar terms used in the present invention do not denote any order, quantity or importance, but are only used to distinguish different components. Words such as "comprising" or "including" mean that the elements or items appearing before this word cover the elements or items listed after this word and their equivalents, without excluding other elements or items. Words such as "connected" or "coupled" are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Upper", "lower", "left", "right", etc. are only used to represent relative positional relationships, and when the absolute position of the object being described changes, the relative positional relationship may also change accordingly.

[0040] Embodiment:

[0041] In some areas that need to be enclosed, such as accident scenes, emergency rescue scenes or production sites, in order to prevent unauthorized personnel from entering, obstacles such as warning lines or fences are usually set up.

[0042] In order to more conveniently monitor whether there are people entering the enclosed area, a more advanced method is to set up an electronic fence. If an intruder touches the electronic fence, an alarm will be sent to the staff so that they can rush to the scene in time to expel the intruder.

[0043] However, for a site with a large enclosed area, due to the large peripheral range of the site, there may be a large number of people around the electronic fence, among which there may be intruders with the intention of intrusion. In order to break through the electronic fence, they will make multiple round trips to the scene. In this case, in order to identify the intruders with the intention of intrusion, using high-alert state monitoring is obviously too labor-consuming. Therefore, there is an urgent need for a method or system that can save labor and effectively prevent intruders.

[0044] Therefore, the present invention discloses a face recognition method for an electronic fence, which is applied with a display module. The method includes:

[0045] Step 1, obtain the video information captured by a monitoring camera, determine the moving targets according to multi-target motion detection technology, and mark the moving targets on the display module.

[0046] Step 2, perform face recognition scanning on the moving targets. If it is determined that there are face features, generate a face feature record table according to the face features, and record the time and frequency of detecting the face features in the face feature record table.

[0047] In step 2, if the time length or frequency of detecting the face feature is greater than a preset value, it is determined that the moving target corresponding to the face feature is a suspected intrusion object, and it is marked on the display module.

[0048] In some embodiments of the present application, in order to be able to determine whether a moving target repeatedly approaches the electronic fence according to the content recorded in the face feature record table, the content of the face record table is disclosed. The content of the face feature record table includes face feature values, several face feature discovery times, and several face feature disappearance times, where the face discovery time and the face feature disappearance time correspond one by one.

[0049] The face feature value is used to distinguish the moving target; the face feature discovery time is used to determine the discovery time when the face feature is discovered; the face feature disappearance time is used to determine the disappearance time of the face feature.

[0050] In some embodiments of the present application, a method for determining the time length of detecting the face feature through the face feature record table is disclosed. The method for determining the time length of detecting the face feature includes:

[0051] Step 3, according to the face feature discovery time and the face feature disappearance time recorded in the face feature record table, determine the time length of the face feature being detected.

[0052] In some embodiments of the present application, a method for determining the frequency of detecting the face feature through the face feature record table is disclosed. The method for determining the frequency of the face feature being detected includes:

[0053] Step 4, according to the update times of the face feature discovery time in the face feature record table recorded within a unit time, determine the frequency of detecting the face feature.

[0054] In some embodiments of the present application, in order to reduce the workload of face recognition scanning and improve the number of moving targets monitored, the method is improved. The improved method is applied before step 1, and the method further includes:

[0055] a. Preset several levels of warning lines outside the electronic fence, and analyze and determine the time length for the moving target to cross adjacent two levels of warning lines.

[0056] b. If the time length for the moving target to cross adjacent two levels of warning lines is less than a preset value, it is determined that the moving target has completed a quick crossing of the warning line.

[0057] c. If the number of times the moving target quickly crosses the warning line is greater than the preset value, then perform face recognition scanning on the moving target again.

[0058] The present application discloses a face recognition method for an electronic fence, which confirms moving targets through multi-target motion detection technology, performs face recognition scanning on the moving targets, generates a face feature record table for the moving targets screened out with face features, and based on the face feature record table, analyzes and determines that the personnel who make multiple round trips or stay for too long are suspected objects with intrusion intentions, avoiding the manual consumption of manual monitoring and not having the problem of omission in manual monitoring.

[0059] In order to further elaborate the technical solution of the present application, the idea steps for constructing a face recognition system for an electronic fence are now disclosed. The requirements for this system are to be able to determine the intruding personnel with intrusion intentions and then alarm the staff to prevent the intruding personnel from breaking through the electronic fence and entering the enclosed site.

[0060] Such as Figure 1 , the idea steps include:

[0061] S1, Obtain the video information captured by the monitoring camera.

[0062] S2, Determine the moving targets according to the multi-target motion detection technology and mark the moving targets on the display module.

[0063] S3, Perform face recognition scanning on the moving targets.

[0064] S4, If it is determined that there are face features, generate a face feature record table according to the face features.

[0065] S5, Record the time and frequency of detecting the face features in the face feature record table.

[0066] S6, If the time length or frequency of detecting the face features is greater than the preset value, determine that the moving target corresponding to the face features is a suspected intrusion object and mark it on the display module.

[0067] In the step S4, the content of the face feature record table includes face feature values, several face feature discovery times, and several face feature disappearance times. The face feature values are used to distinguish the moving targets; the face feature discovery times are used to determine the discovery times of discovering the face features; the face feature disappearance times are used to determine the disappearance times of the face features.

[0068] In the step S6, the method for determining the time length of detecting the face feature includes: determining the time length of detecting the face feature according to the face feature discovery time and the face feature disappearance time recorded in the face feature record table.

[0069] The method for determining the frequency of detecting the face feature includes: determining the frequency of detecting the face feature according to the update times of the face feature discovery time in the face feature record table recorded within a unit time.

[0070] To further improve the execution efficiency of the system, early warning screening can be performed on the moving target, that is, a conceptual step for early warning screening of the moving target. For example Figure 2 , between the step S2 and the step S3, the following conceptual steps are further included:

[0071] S201, Preset several levels of warning lines around the perimeter of the electronic fence.

[0072] S202, Analyze and determine the time length for the moving target to cross adjacent two levels of warning lines.

[0073] S203, If the time length for the moving target to cross adjacent two levels of warning lines is less than the preset value, it is determined that the moving target has completed a quick crossing of the warning line.

[0074] S204, If the number of times the moving target has completed the quick crossing of the warning line is greater than the preset value, then perform a face recognition scan on the moving target again.

[0075] In some embodiments of the present application, a face recognition system for an electronic fence is also disclosed. The system includes a monitoring camera, a display module, and an analysis and processing module.

[0076] The monitoring camera is used to capture the scene of the electronic fence; the display module is used to display the images captured by the monitoring camera; the analysis and processing module is used to analyze and process the video information captured by the monitoring camera.

[0077] The analysis method of the analysis and processing module includes determining the moving target according to the multi-target motion detection technology, and performing a face recognition scan on the moving target. If a face feature is determined to exist, a face feature record table is generated according to the face feature, and the time and frequency of detecting the face feature are recorded in the face feature record table.

[0078] Based on the face feature record table, the analysis and processing module calculates the time length and frequency of the detected face feature, and compares the time length and frequency of the detected face feature with a preset value. If the time length or frequency of the detected face feature is greater than the preset value, it is determined that the moving target corresponding to the face feature is a suspected intrusion object, and the display module is driven to mark the suspected intrusion object.

[0079] In some embodiments of the present application, a method capable of analyzing and determining the time length and frequency of the detected face feature is provided. The method for the analysis and processing module to determine the time length and frequency of the detected face feature includes:

[0080] The content of the face feature record table includes face feature values, several face feature discovery times, and several face feature disappearance times;

[0081] Based on the face feature discovery time and face feature disappearance time recorded in the face feature record table, determine the time length of the detected face feature;

[0082] Based on the update times of the face feature discovery time in the face feature record table recorded within a unit time, determine the frequency of the detected face feature.

[0083] In some embodiments of the present application, in order to reduce the workload of system face recognition scanning and increase the number of monitored moving targets, an early warning module is disclosed to pre-screen and alert the moving target points. The system further includes an early warning module, and the early warning module is used to pre-judge the moving target to determine whether to perform face recognition scanning on the moving target.

[0084] In some embodiments of the present application, a method for the early warning module to pre-screen and judge the moving target is disclosed. The method for the early warning module to pre-screen and judge the moving target includes:

[0085] Preset several levels of warning lines on the periphery of the electronic fence. The early warning module analyzes and determines the time length for the moving target to cross adjacent two levels of warning lines. If the time length for the moving target to cross adjacent two levels of warning lines is less than the preset value, it is determined that the moving target has completed a quick crossing of the warning line. If the number of times the moving target has completed a quick crossing of the warning line is greater than the preset value, the analysis and processing module is enabled to perform face recognition scanning on the moving target.

[0086] The present application discloses a face recognition system for an electronic fence. The on-site shooting is completed through a monitoring camera, the display of the captured image is completed through a display module, and the analysis and confirmation of multiple round-trip personnel and long-term staying personnel are completed through an analysis and processing module, so as to determine the suspected objects with intrusion intentions, avoiding the labor consumption of manual monitoring and the omission problem in the process of manual monitoring, and improving the security of the application of the electronic fence.

[0087] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that they can still modify or equivalently replace the technical solutions of the present invention, and these modifications or equivalent replacements cannot make the modified technical solutions deviate from the spirit and scope of the technical solutions of the present invention.

Claims

1. A face recognition method for an electronic fence, characterized in that, a display module is applied, and the method includes: Obtain the video information captured by the monitoring camera, determine the moving targets according to the multi-target motion detection technology, and mark the moving targets on the display module; Perform face recognition scanning on the moving targets. If it is determined that there are face features, generate a face feature record table according to the face features, and record the time and frequency of detecting the face features in the face feature record table; If the time length or frequency of detecting the face features is greater than the preset value, determine that the moving target corresponding to the face features is a suspected intrusion object, and mark it on the display module; The method further includes: Preset several levels of warning lines outside the electronic fence, and analyze and determine the time length for the moving target to cross adjacent two levels of warning lines; If the time length for the moving target to cross adjacent two levels of warning lines is less than the preset value, it is determined that the moving target has completed a quick crossing of the warning line; If the number of times the moving target has completed a quick crossing of the warning line is greater than the preset value, perform face recognition scanning on the moving target again; The content of the face feature record table includes: Face feature values, used to distinguish the moving target; Several face feature discovery times, used to determine the discovery time of the face features; Several face feature disappearance times, used to determine the disappearance time of the face features; The method for determining the time length of detecting the face features includes: According to the face feature discovery time and face feature disappearance time recorded in the face feature record table, determine the time length of detecting the face features; The method for determining the frequency of detecting the face features includes: According to the update times of the face feature discovery time in the face feature record table recorded within a unit time, determine the frequency of detecting the face features.

2. A face recognition system for an electronic fence, characterized in that, the system includes: A monitoring camera, used to photograph the scene of the electronic fence; A display module, used to display the image captured by the monitoring camera; An analysis and processing module, used to analyze and process the video information captured by the monitoring camera; The analysis method of the analysis and processing module includes determining the moving targets according to the multi-target motion detection technology, and performing face recognition scanning on the moving targets. If it is determined that there are face features, generate a face feature record table according to the face features, and record the time and frequency of detecting the face features in the face feature record table; The analysis and processing module calculates the time length and frequency of detecting the face features according to the face feature record table, and compares the time length and frequency of detecting the face features with the preset value. If the time length or frequency of detecting the face features is greater than the preset value, determine that the moving target corresponding to the face features is a suspected intrusion object, and drive the display module to mark the suspected intrusion object; The system further includes a pre-warning module, which is used to make a preliminary judgment on the moving target to determine whether to perform a face recognition scan on the moving target; The method for the pre-warning module to make a preliminary judgment on the moving target includes: Several levels of warning lines are preset on the periphery of the electronic fence. The pre-warning module analyzes and determines the time length for the moving target to cross two adjacent warning lines. If the time length for the moving target to cross two adjacent warning lines is less than a preset value, it is determined that the moving target has completed a rapid crossing of the warning line. If the number of times the moving target has completed a rapid crossing of the warning line is greater than a preset value, the analysis and processing module is enabled to perform a face recognition scan on the moving target; The method for the analysis and processing module to determine the time length and frequency of detecting the face feature includes: The content of the face feature record table includes face feature values, several face feature discovery times, and several face feature disappearance times; According to the face feature discovery time and face feature disappearance time recorded in the face feature record table, determine the time length of detecting the face feature; According to the update times of the face feature discovery time in the face feature record table recorded within a unit time, determine the frequency of detecting the face feature.

Images