Intrusion guard method and device based on scene recognition

Abstract

The invention belongs to the field of security and protection, and particularly relates to an intrusion guarding method and device based on scene recognition. The method comprises the following steps: S1, acquiring a pedestrian image through a camera; s2, determining whether the pedestrian is a stranger or not according to the pedestrian image, and if yes, performing first-level security and protection processing including but not limited to pedestrian behavior recording; s3, determining whether a stranger enters a thermodynamic area according to the position of the stranger in the pedestrian image, and if yes, performing secondary security and protection processing including but not limited to starting of an alarm; and S4, according to the position of a stranger in the pedestrian image, determining whether the stranger leaves or enters the intrusion protection area at the specified intrusion boundary of the thermodynamic area, and if so, performing three-level security and protection processing including but not limited to urgent contact with the user. According to the invention, the recognition accuracy of pedestrian behaviors is improved through different security levels, and the security credibility is improved.

Description

Technical Field

[0001] The present application relates to the field of security, and in particular to an intrusion protection method and device based on scene recognition. Background Art

[0002] Today's security focuses more on property protection, but there is no particularly accurate solution in the existing technology. Although there is face or behavior recognition, the detection rate cannot be improved due to some objective reasons such as the camera installation angle and distance. At the same time, it will also cause security manufacturers to be afraid to be too aggressive when dealing with this gray-scale sensitive but user-focused issue, fearing false alarms, which in turn causes security anxiety. Summary of the invention

[0003] In order to solve the above technical problems, the present application provides an intrusion protection method and device based on scene recognition, which identifies and statistically distinguishes scene information, increases the weight of scene blocks, and performs security processing at different levels.

[0004] The first aspect of the present application provides an intrusion protection method based on scene recognition, which mainly includes: Step S1, obtaining a pedestrian image through a camera; Step S2: determining whether the pedestrian is a stranger based on the pedestrian image, and if so, performing primary security processing, including but not limited to recording the pedestrian's behavior; Step S3, determining whether the stranger has entered the thermal area according to the position of the stranger in the pedestrian image, and if so, performing secondary security processing, including but not limited to activating an alarm; Step S4: determine whether the stranger leaves or enters the intrusion protection area at the designated intrusion boundary of the thermal area according to the position of the stranger in the pedestrian image. If so, perform level 3 security processing, including but not limited to emergency contacting the user.

[0005] Preferably, before step S3, the method further comprises: Step S31: within a set time period, count the number of pedestrians passing by each pixel point in the camera shooting area as the thermal value of the pixel point; Step S32, constructing a thermal path according to the thermal value and the thermal threshold; Step S33: taking the thermal path as the thermal area.

[0006] Preferably, step S31 further comprises: The number of pedestrians passing each pixel is added with a time point label, and the time points where the number exceeds the set value are extracted to form a time period. The pedestrian behavior in the time period is set to be processed using no more than level 2 security.

[0007] Preferably, before step S4, the method further comprises: The user specifies an intrusion protection point, and forms an area of ​​a specified shape with the intrusion protection point as the center as the intrusion protection area, or the user specifies a partial boundary line segment of the thermal area close to the property to be protected as the intrusion boundary.

[0008] The second aspect of the present application provides an intrusion protection device based on scene recognition, which mainly includes: A pedestrian image acquisition module is used to acquire pedestrian images through a camera; A primary security processing module, used to determine whether the pedestrian is a stranger based on the pedestrian image, and if so, perform primary security processing, including but not limited to recording the pedestrian's behavior; A secondary security processing module, used to determine whether a stranger has entered the thermal area according to the position of the stranger in the pedestrian image, and if so, perform secondary security processing, including but not limited to activating an alarm; The third-level security processing module is used to determine whether the stranger leaves or enters the intrusion protection area at the designated intrusion boundary of the thermal area based on the position of the stranger in the pedestrian image. If so, the third-level security processing is performed, including but not limited to emergency contacting the user.

[0009] Preferably, the intrusion protection device further comprises: The thermal value statistics unit is used to count the number of pedestrians passing by each pixel point in the camera shooting area within a set time period as the thermal value of the pixel point; A thermal path construction unit, used for constructing a thermal path according to a thermal value and a thermal threshold; The thermal zone generation unit is used to use the thermal path as the thermal zone.

[0010] Preferably, the thermal value statistical unit includes: The time limitation subunit is used to add a time point label to the number of pedestrians passing each pixel, extract the time points where the number exceeds the set value, form a time period, and set the pedestrian behavior within the time period to be processed using no more than level 2 security.

[0011] Preferably, the intrusion protection device further comprises: The intrusion boundary designation module is used for the user to designate an intrusion protection point, and to form an area of ​​a designated shape with the intrusion protection point as the center as the intrusion protection area, or for the user to designate a partial boundary line segment of the thermal area close to the property to be protected as the intrusion boundary.

[0012] This application improves the accuracy of pedestrian behavior recognition and enhances security credibility through different security levels. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 It is a flowchart of a preferred embodiment of the intrusion protection method based on scene recognition of the present application. DETAILED DESCRIPTION

[0014] In order to make the purpose, technical scheme and advantages of the implementation of this application clearer, the technical scheme in the implementation of this application will be described in more detail in combination with the drawings in the implementation of this application. In the drawings, the same or similar reference numerals throughout represent the same or similar elements or elements with the same or similar functions. The described implementation is a part of the implementation of this application, not all of the implementations. The implementation described below with reference to the drawings is exemplary and is intended to be used to explain this application, and cannot be understood as a limitation on this application. Based on the implementation in this application, all other implementations obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application. The implementation of this application is described in detail below in combination with the drawings.

[0015] The first aspect of the present application provides an intrusion protection method based on scene recognition, such as Figure 1 As shown, it mainly includes: Step S1: Acquire pedestrian images through a camera.

[0016] Step S2: Determine whether the pedestrian is a stranger based on the pedestrian image. If yes, perform primary security processing, including but not limited to recording the pedestrian's behavior.

[0017] Step S2 of the present application provides the lowest level of security protection, mainly for situations such as strangers appearing in the camera shooting area. The system records these routine activities or makes ordinary notifications, and does not automatically trigger an alarm. Users can view and share them on the terminal as needed.

[0018] Step S3: determine whether a stranger has entered the thermal area according to the position of the stranger in the pedestrian image. If yes, perform secondary security processing, including but not limited to activating an alarm.

[0019] This step S3 is mainly aimed at strangers entering the thermal area. The thermal area is usually the user's private area. The entry of strangers into this area usually indicates that there are safety hazards or dangerous things. The security level should be increased and secondary security processing should be carried out. Secondary security is based on primary security and adds alarm warning notifications. For example, an alarm is generated in the security area, or an alarm is generated on the user terminal to notify the user to preview.

[0020] In some optional implementations, the process further includes before step S3: Step S31: within a set time period, count the number of pedestrians passing by each pixel point in the camera shooting area as the thermal value of the pixel point; Step S32, constructing a thermal path according to the thermal value and the thermal threshold; Step S33: taking the thermal path as the thermal area.

[0021] This embodiment provides one of the methods for constructing a thermal area, that is, by constructing a thermal path and using the thermal path as a thermal area. The reason for constructing a thermal path is usually because the camera cannot directly face the intrusion protection area such as the property to be protected or doors and windows. For example, for home monitoring, the scene monitoring screen cannot see information such as doors and windows, only the roads in the courtyard. At this time, the thermal path in the monitoring screen can be constructed. In this embodiment, the scene information within a period of time is counted, and each time the trajectory of the character's movement path contains a number of picture pixels. The thermal information of each pixel is counted, and then according to the preset layered thermal threshold, after a period of time, the entire scene screen can be formed into a thermal path according to the layered threshold. At the same time, there will also be a corresponding cold path. For example, the path from the courtyard gate to the front of the house or to the side door is a thermal path, and other areas such as lawns are cold paths.

[0022] In some optional implementations, step S31 further includes: The number of pedestrians passing each pixel is added with a time point label, and the time points where the number exceeds the set value are extracted to form a time period. The pedestrian behavior in the time period is set to be processed using no more than level 2 security.

[0023] This embodiment mainly counts the time patterns of pedestrians passing through the thermal path. For example, an event of leaving early and returning late can be connected in series through these two points + time information. When leaving early in the morning, the event path is to first pass through the main entrance point of the house, and then pass through the yard entrance point. When returning at night, first pass through the yard entrance point, and then pass through the main entrance point of the house. Generally speaking, during this time period, since the householder often uses this thermal path, even if a stranger passes by, usually no property loss will occur. It is more likely that the stranger is accompanying or visiting. At this time, only the secondary security processing can be used, and there is no need to increase the security level in subsequent steps.

[0024] Step S4: determine whether the stranger leaves or enters the intrusion protection area at the designated intrusion boundary of the thermal area according to the position of the stranger in the pedestrian image. If so, perform level 3 security processing, including but not limited to emergency contacting the user.

[0025] This step is mainly for strangers entering the intrusion protection area. At this time, the highest level of security protection is adopted, such as the fastest and highest priority emergency notification to the user. At the same time, you can also set in advance whether to enable the alarm warning. It includes two situations. One is that the camera happens to be able to capture the intrusion protection area. For example, the scene monitoring screen can see information such as doors and windows. In combination with the picture recognition, it automatically makes the protection point of the scene. Another situation is that the camera cannot capture the intrusion protection area. At this time, it can be estimated whether the stranger will enter the intrusion protection area based on the position where the stranger walks out of the camera collection range along the thermal area. When the stranger walks along the thermal area, there are usually two situations of being far away from or close to the intrusion protection area. The camera boundary close to the intrusion protection area is defined as the intrusion boundary. When the stranger walks through the intrusion boundary, the security level can be increased to level three security.

[0026] In some optional implementations, the process further includes before step S4: The user specifies an intrusion protection point, and forms an area of ​​a specified shape with the intrusion protection point as the center as the intrusion protection area, or the user specifies a partial boundary line segment of the thermal area close to the property to be protected as the intrusion boundary.

[0027] This embodiment provides a method for constructing an intrusion protection area. Within the camera acquisition area, the user usually first specifies the property to be protected or the critical path as an intrusion protection point. Then, a protection area is automatically generated in the surrounding area of ​​the intrusion protection point (for example, a square area of ​​100*100 pixels, which can be customized according to the screen ratio) as the intrusion protection area. Due to the camera installation angle problem, for the property to be protected that cannot be photographed, the intrusion boundary is determined by delineating the boundary line segment of the camera.

[0028] Through the technical solution of the present application, the inaccuracy of security event notifications is avoided, so that users can obtain the things they want to know most, care about most, or are important, thereby improving the user experience.

[0029] The second aspect of the present application provides an intrusion protection device based on scene recognition corresponding to the above method, mainly comprising: A pedestrian image acquisition module is used to acquire pedestrian images through a camera; A primary security processing module, used to determine whether the pedestrian is a stranger based on the pedestrian image, and if so, perform primary security processing, including but not limited to recording the pedestrian's behavior; A secondary security processing module, used to determine whether a stranger has entered the thermal area according to the position of the stranger in the pedestrian image, and if so, perform secondary security processing, including but not limited to activating an alarm; The third-level security processing module is used to determine whether the stranger leaves or enters the intrusion protection area at the designated intrusion boundary of the thermal area based on the position of the stranger in the pedestrian image. If so, the third-level security processing is performed, including but not limited to emergency contacting the user.

[0030] In some optional implementations, the intrusion protection device further includes: The thermal value statistics unit is used to count the number of pedestrians passing by each pixel point in the camera shooting area within a set time period as the thermal value of the pixel point; A thermal path construction unit, used for constructing a thermal path according to a thermal value and a thermal threshold; The thermal zone generation unit is used to use the thermal path as the thermal zone.

[0031] In some optional implementations, the thermal value statistical unit includes: The time limitation subunit is used to add a time point label to the number of pedestrians passing each pixel, extract the time points where the number exceeds the set value, form a time period, and set the pedestrian behavior within the time period to be processed using no more than level 2 security.

[0032] In some optional implementations, the intrusion protection device further includes: The intrusion boundary designation module is used for the user to designate an intrusion protection point, and to form an area of ​​a designated shape with the intrusion protection point as the center as the intrusion protection area, or for the user to designate a partial boundary line segment of the thermal area close to the property to be protected as the intrusion boundary.

[0033] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application shall be based on the protection scope of the claims.

Claims

1. An intrusion protection method based on scene recognition, characterized in that: include: Step S1, obtaining a pedestrian image through a camera; Step S2: determining whether the pedestrian is a stranger based on the pedestrian image, and if so, performing primary security processing, including but not limited to recording the pedestrian's behavior; Step S3, determining whether the stranger has entered the thermal area according to the position of the stranger in the pedestrian image, and if so, performing secondary security processing, including but not limited to activating an alarm; Step S4: determine whether the stranger leaves or enters the intrusion protection area at the designated intrusion boundary of the thermal area according to the position of the stranger in the pedestrian image. If so, perform level 3 security processing, including but not limited to emergency contacting the user.

2. The intrusion protection method based on scene recognition as claimed in claim 1, characterized in that: Before step S3, the method further comprises: Step S31: within a set time period, count the number of pedestrians passing by each pixel point in the camera shooting area as the thermal value of the pixel point; Step S32, constructing a thermal path according to the thermal value and the thermal threshold; Step S33: taking the thermal path as the thermal area.

3. The intrusion protection method based on scene recognition as claimed in claim 2, characterized in that: Step S31 further includes: The number of pedestrians passing each pixel is added with a time point label, and the time points where the number exceeds the set value are extracted to form a time period. The pedestrian behavior in the time period is set to be processed using no more than level 2 security.

4. The intrusion protection method based on scene recognition as claimed in claim 1, characterized in that: Before step S4, the method further comprises: The user specifies an intrusion protection point, and forms an area of ​​a specified shape with the intrusion protection point as the center as the intrusion protection area, or the user specifies a partial boundary line segment of the thermal area close to the property to be protected as the intrusion boundary.

5. An intrusion protection device based on scene recognition, characterized in that: include: A pedestrian image acquisition module is used to acquire pedestrian images through a camera; A primary security processing module, used to determine whether the pedestrian is a stranger based on the pedestrian image, and if so, perform primary security processing, including but not limited to recording the pedestrian's behavior; A secondary security processing module, used to determine whether a stranger has entered the thermal area according to the position of the stranger in the pedestrian image, and if so, perform secondary security processing, including but not limited to activating an alarm; The third-level security processing module is used to determine whether the stranger leaves or enters the intrusion protection area at the designated intrusion boundary of the thermal area based on the position of the stranger in the pedestrian image. If so, the third-level security processing is performed, including but not limited to emergency contacting the user.

6. The intrusion protection device based on scene recognition as claimed in claim 5, characterized in that: The intrusion protection device also includes: The thermal value statistics unit is used to count the number of pedestrians passing by each pixel point in the camera shooting area within a set time period as the thermal value of the pixel point; A thermal path construction unit, used for constructing a thermal path according to a thermal value and a thermal threshold; The thermal zone generation unit is used to use the thermal path as the thermal zone.

7. The intrusion protection device based on scene recognition as claimed in claim 6, characterized in that: The thermal value statistical unit includes: The time limitation subunit is used to add a time point label to the number of pedestrians passing each pixel, extract the time points where the number exceeds the set value, form a time period, and set the pedestrian behavior within the time period to be processed using no more than level 2 security.

8. The intrusion protection device based on scene recognition as claimed in claim 5, characterized in that: The intrusion protection device also includes: The intrusion boundary designation module is used for the user to designate an intrusion protection point, and to form an area of ​​a designated shape with the intrusion protection point as the center as the intrusion protection area, or for the user to designate a partial boundary line segment of the thermal area close to the property to be protected as the intrusion boundary.

Images