A method for target authorization based on continuous radar detection and tracking, a method for intrusion detection, and corresponding electronic equipment
Through radar continuous detection and tracking technology, combining distance and motion direction to determine target permissions, privacy and false alarm problems in the existing technology are solved, and efficient intrusion detection and authorization in industrial and residential scenarios are achieved.
Patent Information
- Application Number
- CN202010981541.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2020-09-17
- Publication Date
- 2024-12-27
- Estimated Expiration
- 2040-09-17
AI Technical Summary
Among the existing intrusion detection technology, there are privacy problems based on image recognition technology, which is difficult to distinguish target permissions, and voiceprint recognition is difficult to obtain target location information, which is easy to cause false alarms.
The method based on radar continuous detection and tracking is adopted, by obtaining prevention and control area information, continuously collecting point cloud information, detecting and authorizing targets, and using parameters such as distance and motion direction to determine target permissions, so as to achieve authorization and intrusion detection of unauthorized targets.
It realizes the accurate distinction between target permissions without infringing on user privacy, reduces false alarms, improves the accuracy and security of intrusion detection, and is suitable for multi-region monitoring in complex scenarios.
Smart Images

Figure CN114200443B_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field of security, and relates to a method for target authorization based on continuous radar detection and tracking, a method for intrusion detection based on continuous radar detection and tracking, and corresponding electronic equipment. Background Art
[0002] With the advancement of technology and the continuous improvement of people's safety awareness, it is very common to install intrusion detection systems in industrial, commercial and residential areas. The detection and identification of intruders can adopt a variety of different or combined technical means, such as vibrating optical fiber, image recognition, sound recognition, etc. For example, patent No. CN102521578B provides a method for intrusion detection based on image recognition technology, which first extracts a representative frame image from a real-time video, and then successively performs human target recognition, face recognition, manual recognition, etc. to accurately judge strangers. Patent No. CN111311860A provides a method for intrusion detection by collecting and comparing footsteps signals, including: collecting footsteps in the area to be detected to obtain footsteps signals; extracting the voiceprint features of the footsteps signal as the voiceprint features to be verified; based on the similarity between the voiceprint features to be verified and the authenticated voiceprint features, determining whether the footsteps are the footsteps of an authorized person, and the authenticated voiceprint features are the voiceprint features extracted from the footsteps signals of the authenticated person; if the footsteps are not the footsteps of the authorized person, determining that the area to be detected has an area intrusion. These methods have their own advantages and disadvantages. Among them, intrusion detection based on image recognition technology involves privacy issues and is therefore not widely accepted by the majority of users, especially residential users. Voiceprint recognition is difficult to obtain the target's location information and is easily affected by environmental interference in soft isolation boundary scenarios. It is difficult to confirm whether a target appearing nearby is actually within the prevention and control area, which can easily cause false alarms.
[0003] The above problems can be avoided by using radar to transmit electromagnetic wave signals to the control area and receive echoes, thereby identifying and tracking the intruding target. However, radar is not as intuitive as image recognition technology and cannot use biometric information such as face and iris to judge the target type. Therefore, when using radar technology for intrusion detection, how to distinguish target permissions is particularly critical. Summary of the invention
[0004] The purpose of the present invention is to provide a method for target authorization based on continuous radar detection and tracking and corresponding electronic equipment, and secondly, to provide a method for intrusion detection based on continuous radar detection and tracking and corresponding electronic equipment.
[0005] To achieve the above purpose, the present invention adopts the following technical means:
[0006] First aspect:
[0007] A method for target authorization based on continuous radar detection and tracking, the steps comprising:
[0008] S1: Acquire control area information, where the control area information at least includes control area boundary information and entry authorization area information;
[0009] S2: Continuously collect point cloud information of active targets in the prevention and control area, and detect and track active targets;
[0010] S3: Detect and identify unauthorized new targets in the control area. When an unauthorized new target is detected in the entry authorization area, a waiting time is given to the new target, and the target authorization status is updated to pending; the distance between the pending target and each authorized target at each collection time during the waiting time is traversed to determine whether there is a distance less than or equal to the set threshold. If so, the pending target is authorized and the target authorization status is updated to authorized.
[0011] In step S3, the distance between the pending target and each authorized target and the movement direction of the authorized target at each collection moment during the waiting time are traversed to determine whether there is an authorized target whose movement direction points to the pending target and the distance between the authorized target and the pending target is less than or equal to the set threshold. If so, the pending target is authorized.
[0012] If the angle between the line connecting the positions of the authorized target and the pending target and the moving direction of the authorized target is less than or equal to a set threshold, it is determined that the moving direction of the authorized target points to the pending target.
[0013] There are multiple control areas, and the monitoring ranges of radar monitoring equipment deployed in two control areas overlap to form a boundary. The two control areas are adjacent to each other, and the formed boundary is an adjacent authorization area. The adjacent authorization area is an entry authorization area shared by the two control areas; multiple control areas connected by adjacent authorization areas form a joint control area; when an authorized target is tracked to move from one control area to an adjacent authorization area, the target is given authorization to be associated with the adjacent control area of the adjacent authorization area.
[0014] Second aspect:
[0015] An electronic device comprises: one or more processors; a storage device on which one or more programs are stored; when the one or more programs are executed by the one or more processors, the one or more processors implement the method described in the first aspect.
[0016] The third aspect:
[0017] A method for intrusion detection based on continuous radar detection and tracking, the steps comprising:
[0018] S10: Acquire control area information, where the control area information at least includes control area boundary data and entry authorization area information;
[0019] S20: Continuously collect point cloud information of active targets in the prevention and control area, and detect and track active targets;
[0020] S30: Detect and identify unauthorized new targets, and determine whether the first appearance position of the new target is within any entry authorization area; if not, determine that the target is an intruder; if so, give the target a waiting time and mark the target as a pending target;
[0021] S40: Check whether the pending target is authorized through any of the following authorization methods within the waiting time: A collects the identification information of the pending target that can be used to obtain authorization, verifies and identifies the identification information, and if it matches, the pending target is authorized and marked as an authorized target; B traverses the distance between the pending target and the authorized target at each collection moment, and determines whether there is a distance less than or equal to a set threshold. If so, the pending target is authorized and marked as an authorized target; if the pending target still has not been authorized after the waiting time expires, the authorization of the target is explicitly rejected, and the target is determined to be an intruder.
[0022] S50: When the target is tracked to leave the current control area and exceeds the set time, the information of the target is cleared and the tracking of the target is stopped.
[0023] There are multiple control areas, and the monitoring ranges of radar monitoring equipment deployed in two control areas overlap to form a boundary. The two control areas are adjacent to each other, and the boundary formed is an adjacent authorization area. The adjacent authorization area is an entry authorization area shared by the two control areas; multiple control areas connected by adjacent authorization areas form a joint control area; the authorization method of step S40 also includes C adjacent area authorization: when an authorized target is tracked to move from one control area to an adjacent authorization area, the target is given authorization to be associated with the adjacent control area of the adjacent authorization area.
[0024] The authorization method also includes D compulsory authorization: receiving a compulsory authorization instruction issued by a user and granting compulsory authorization to the target.
[0025] In the authorization method B of step S40, the distance between the pending target and each authorized target and the movement direction of the authorized target at each collection moment during the waiting time are traversed to determine whether there is an authorized target whose movement direction points to the pending target and the distance between the authorized target and the pending target is less than or equal to the set threshold. If so, the pending target is authorized.
[0026] If the angle between the line connecting the positions of the authorized target and the pending target and the moving direction of the authorized target is less than or equal to the set threshold, the moving direction of the authorized target is deemed to be pointing to the pending target.
[0027] Step S50 specifically includes: traversing the point cloud information of each target collected for the last time, and determining whether the difference between the collection time and the current time exceeds the set threshold; if so, determining whether the position of the collected target is within the entry authorization area or within the set threshold range near the boundary of the control area; if so, determining that the target has left the current control area and exceeded the set time, clearing the target's information and stopping tracking of the target.
[0028] When a target is determined to be an intruder, an alarm message is output.
[0029] The identification information includes the target's biometric information, the electronic identification carried by the target, and the electronic identification generated by the smart door lock when the target uses a legitimate key to open the door.
[0030] The biometric information includes a person's face, iris, fingerprint, voiceprint, posture, and gesture information. The verification and recognition of biometric information can be achieved by training a verification and recognition model using a machine learning method.
[0031] The electronic identification includes an electronic identification based on RFID, BLE, mobile terminal MAC, mobile terminal APP key entry and / or NFC technology.
[0032] Fourth aspect:
[0033] An electronic device comprises: one or more processors; a storage device on which one or more programs are stored; when the one or more programs are executed by the one or more processors, the one or more processors implement the method described in the third aspect.
[0034] In the first and third aspects, the point cloud information includes at least the acquisition time and the position of the target at that time. As a preferred solution, the point cloud information includes the acquisition time and the position, speed, and acceleration of the target at that time. When collecting point cloud information, the radar emits electromagnetic waves to the control area at a frequency of 5 to 200 times per second. The frequency range of the electromagnetic waves is 4G to 120G, preferably 20 to 100G. In the 20 to 100G frequency band, electromagnetic waves will be blocked by walls, and the radar coverage range is consistent with the range circled by the wall building.
[0035] Target identification is performed based on radar detection and tracking. Compared with infrared technology, radar equipment is easy to install, not affected by light intensity, and has a wide deployment range and no blind spots; compared with image technologies such as video surveillance, radar radio frequency signals are not affected by light, smoke and fog, and there is no privacy leakage. Radar has a natural advantage in obtaining target position and trajectory, but unlike image technology, it can not directly obtain the target's facial, iris and other biological information to easily distinguish the target. Of course, the purpose of distinguishing targets can also be achieved by using point cloud to identify people's postures, but this method is relatively complex to calculate and has high requirements for equipment. The authorization method provided by the present invention is designed based on position and trajectory, and responds to authorized targets (such as owners in residential scenes, internal employees in industrial scenes, security managers and other resident personnel) to open the door or greet unauthorized targets (visitors) for authorization. It is particularly suitable for scenarios where the right of alert can be returned to the resident personnel as long as the resident personnel notice the visitor, and can achieve non-perceptual authorization of visitors, and the user experience is good. When the prevention and control area is large or the divisions are complex and require multiple sets of equipment for zone monitoring, each zone can be monitored and calculated separately as an independent area, which can effectively reduce the amount of calculation. In conjunction with the authorization method for adjacent areas, authorization is granted to the target adjacent area based on the associated area, which simplifies the authorization and authorization transfer process, and obtains security protection without sacrificing convenience.
[0036] The intrusion detection method provided by the present invention is based on radar detection and tracking technology. The equipment is easy to install and has strong environmental adaptability. During the monitoring process, the user is almost not required to take active cooperation actions, and it helps to protect user privacy. Multiple authorization methods are combined for fine-grained authorization, which obtains security protection without sacrificing convenience and reduces the embarrassment of false alarms. BRIEF DESCRIPTION OF THE DRAWINGS
[0037] Figure 1 Shown is a schematic diagram of the deployment of independent control area equipment and its entry authorization area;
[0038] Figure 2 It is a schematic diagram of the deployment of equipment in the joint prevention and control area and the entry authorization area. DETAILED DESCRIPTION
[0039] The present application is further described below in conjunction with the accompanying drawings and specific embodiments.
[0040] Example 1
[0041] A method for target authorization based on continuous radar detection and tracking, the steps comprising:
[0042] S1: Obtaining control area information, the control area information at least includes control area boundary information and entry authorization area information; Figure 1Taking the rectangular independent control area 2 shown in the figure as an example, the radar 1 used to collect point cloud information of active targets in the control area is deployed in a corner area at the top of the control area, and a coordinate system is established with the corner fixed point as the origin. The boundary information of the control area can be represented by the side length of the rectangular area, or by multiple detection points and connecting each detection point to form a domain boundary. For the convenience of calculation, the entry authorization area 3 is represented by one or more circles, and the entry authorization area information includes the center position and radius of the one or more circles. Figure 1 As shown, the entry authorization area 3 of the rectangular independent control area 2 is ( P 1 , r 1 )、( P 2 ,r 2 )、( P 3 ,r 3 )The area covered by these three circular surfaces.
[0043] S2: Continuously collect point cloud information of active targets in the prevention and control area, and detect and track active targets.
[0044] S3: Detect and identify unauthorized new targets in the control area. When an unauthorized new target is detected in the entry authorization area, a waiting time is given to the new target, and the target authorization status is updated to pending; the distance between the pending target and each authorized target at each collection time during the waiting time is traversed to determine whether there is a distance less than or equal to the set threshold. If so, the pending target is authorized and the target authorization status is updated to authorized.
[0045] In the above implementation, only the distance between the pending target and each authorized target is used as the basis for judgment. The smaller the threshold of the distance is set, the safer it is, for example, within 1m. If the threshold is set too large, the authorization intention may be misjudged, resulting in erroneous authorization.
[0046] As a preferred implementation, the moving direction of the authorized target can be further used as an auxiliary basis for judgment. That is, in step S3, the distance between the pending target and each authorized target at each acquisition moment during the waiting time and the moving direction of the authorized target are traversed to determine whether there is an authorized target whose moving direction points to the pending target and the distance between the authorized target and the pending target is less than or equal to the set threshold. If so, the pending target is authorized. If the angle between the line connecting the positions of the authorized target and the pending target and the moving direction of the authorized target is less than or equal to the set threshold, it is determined that the moving direction of the authorized target points to the pending target.
[0047] Example 2
[0048] A method for target authorization based on continuous radar detection and tracking, suitable for scenarios with multiple control areas. If the monitoring ranges of radar monitoring equipment deployed in two control areas overlap to form a boundary, the two control areas are adjacent to each other, and the boundary formed is an adjacent authorization area. The adjacent authorization area is an entry authorization area shared by the two control areas; multiple control areas connected by adjacent authorization areas form a joint control area;
[0049] The steps include:
[0050] S1: Acquire control area information, which includes boundary information of each control area, information of each entry authorization area, and information of adjacent authorization areas. Figure 2 A schematic diagram of the deployment of a joint control area 2, radar equipment 1 and entry authorization area is shown. Each control area can establish a coordinate system independently or calculate in the form of a global coordinate system. Regardless of the form of representation, the mutual conversion between coordinate systems is very easy to implement for those skilled in the art. For the convenience of calculation, the entry authorization area 3 and the adjacent authorization area 4 are represented by one or more circles. The information of these authorization areas includes the center position and radius of the corresponding one or more circles. Figure 2 As shown, ( P 1 ,r 1 )、( P 2 ,r 2 )、( P 3 ,r 3 ) The area covered by these three circles is an entry authorization area 3, the circle ( P 4 ,r 4 ) covers an area that is a neighboring authorized area 4, ( P 5 ,r 5 )、( P 6 ,r 6 )、( P 7 ,r 7 )、( P 8 ,r 8 ) The area covered by these four circular surfaces is an adjacent authorized area 4.
[0051] S2: Radar equipment deployed in each control area collects point cloud information of active targets in its respective control area, and detects and tracks the active targets.
[0052] S3: Detect and identify new unauthorized targets in the control area. When an unauthorized new target is detected in the entry authorization area, a waiting time is given to the new target, and the target authorization status is updated to pending; the distances between the pending target and the authorized targets at each collection time during the waiting time are traversed to determine whether there is a distance less than or equal to the set threshold. If so, the pending target is authorized and the target authorization status is updated to authorized; when an authorized target is tracked to move from one control area to an adjacent authorization area, the target is authorized to be associated with the adjacent control area of the adjacent authorization area, and the target status in the adjacent area is updated to authorized.
[0053] Example 3
[0054] A method for intrusion detection based on continuous radar detection and tracking, the steps comprising:
[0055] S10: Obtaining control area information: The control area can be an independent area or a joint control area formed by multiple control areas connected by adjacent authorized areas. An adjacent authorized area means that when the monitoring ranges of radar monitoring equipment deployed in two control areas overlap to form a boundary, the two control areas are adjacent to each other, and the boundary formed is an adjacent authorized area. The adjacent authorized area is an entry authorization area shared by the two control areas. For an independent control area, the control area information includes control area boundary data and entry authorization area information. For a joint control area, the control area information includes each control area boundary information, each entry authorization area information, and adjacent authorization area information.
[0056] S20: Continuously collect point cloud information of active targets in the prevention and control area, and detect and track active targets.
[0057] S30: Detect and identify unauthorized new targets, and determine whether the first appearance position of the new target is within any entry authorization area; if not, determine that the target is an intruder; if so, give the target a waiting time and mark the target as a pending target.
[0058] S40: Check whether the pending target is authorized through any of the following authorization methods within the waiting time:
[0059] A collects identification information of pending targets that can be used to obtain authorization. For example, a camera installed at the door is used to collect facial information of a person, or a smart door lock is used to collect the target's fingerprint, password, etc., or an electronic identification such as an access card carried by the target, and the identification information is verified and identified. If a match is found, the pending target is authorized and marked as an authorized target. When authorization is performed using method A, a group of targets that are close enough are not distinguished, that is, people traveling together can obtain authorization together.
[0060] B traverses the distance between the pending target and the authorized target at each collection moment, and determines whether there is a distance less than or equal to a set threshold. If so, the pending target is authorized and marked as an authorized target.
[0061] C Adjacent area authorization: When an authorized target is tracked to move from one control area to an adjacent authorized area, the target is authorized to be associated with the adjacent control area of the adjacent authorized area.
[0062] D Forced Authorization: Receive the forced authorization instruction issued by the user and grant forced authorization to the target.
[0063] If the pending target still has not been authorized after the waiting time expires, the authorization of the target is explicitly denied and the target is determined to be an intruder.
[0064] S50: Traverse the point cloud information of each target that was collected the last time, and determine whether the difference between the collection time and the current time exceeds the set threshold; if it exceeds, determine whether the position of the collected target is within the entry authorization area or within the set threshold range near the boundary of the control area. If so, determine that the target has left the current control area and exceeded the set time, clear the target information and stop tracking the target.
[0065] During the execution of the above steps, when a target is determined to be an intruder, an alarm message is output; after the user receives the alarm message, he or she can issue a forced authorization command through gestures, voice, collection APP, etc. to give the target forced authorization.
[0066] The following is a more specific implementation process of the above method:
[0067] 1. Installation and deployment of equipment and configuration: list of equipment in the prevention and control area N =( R q |R=(IP,Bluetooth,P,Size,θ ) ,1≤q ≤ Total number of devices ),in R For device information, including the device IP Address and / or Bluetooth Address (To obtain more information, this embodiment will IP Address and BluetoothThe address is also used as device information), spatial location information P (The spatial position coordinates of the corresponding device in this list ( X, Y ), the size of the control area (for example, the side length of a rectangular control area), the angle between the device coordinate axis and the space coordinate axis θ ; The admission authorization area is represented by the area center point and radius, and the admission authorization area list S ={( O m ,P m ,r m )|0≤ m ≤ Total number of authorized entry areas}, where O For the devices in this area N Index (starting from 0), spatial location information P The coordinates of the center point of the entry authorization area in this list ( X, Y ), r is the radius of the entry authorization area. P and direction θ Both use a two-dimensional coordinate system, such as Figure 1 and Figure 2 As shown, the actual scene can be two-dimensional or three-dimensional.
[0068] 2. Start system initialization parameters: target list LT ={ T n |T=(time, position, velocity, acceleration),I≥n≥ 0}, T is the tracked target, time is the time when the target information was last obtained; Target authorization status table A ={ a n |a ∈( 0 Not authorized, 1 To be determined, 2 Authorized, 3 reject) , I ≥ n ≥ 0};counter I , used to record the number of all targets that have appeared in the scene; target number list D ={ d n | d, I≥d≥1, I≥n≥0}, d Number the target; list of pending targets E ={( d n ,e n )| I≥ d≥1,n≥0 )}, eThe time when the target is added to the pending target list; the adjacent area authorization list LZ= {( s f ,e f ) |s≥0,f ≥0}, s is the region index, e Expiration time; mandatory authorization list LG= { e k |k≥0}, e The expiration time. Set LT , A , D , E , LZ , LG is empty, I is 0.
[0069] 3. Continuous scanning scenario: The device emits electromagnetic waves to the control area at a frequency of more than 5 times per second, detects active targets (human beings, vehicles or other moving objects) in the control area through point cloud, and obtains the current target list C ={ T n |T,n≥ 0}.
[0070] 4. Continue to track targets within the prevention and control area: C Each element in C i , traverse LT Each target in LT j ,contrast C i and LT j of T Whether the attribute values (such as a combination of distance, speed, and time) differ from each other within a certain given value range. If so, C i Information updated to LT j If it exceeds the given range, LT A new record is added in C i Contains T Information, counter I Increase by 1, A Add a record with a value of 0. D Add a value to update I value records.
[0071] 5. Clean up old targets that have left the control area: traversal LT All elements in LT i of T If the time difference between the information and the current time exceeds the set threshold (such as 10 minutes) and the location is within the entry authorization area or within the set threshold range near the control area boundary, it is determined that the target has left the current control area and exceeded the set time, and the index value is i The elements from the corresponding LT , A , D , E Delete it.
[0072] , For the target in the entry authorization area, extend the waiting time and give it a waiting time: traverse A i =0 Each unauthorized element of i Get LT i , traverse each O≠0 Authorized entry area S j ,judge LT i Whether the location attribute is in any of the entry authorization area lists ( O,P,r ) range, if so, update A i =1 "To be determined", and the current time and D i Add to pending list E middle.
[0073] 7. Explicitly deny authorization for targets that exceed the target limit: Take out each index as i and A i =1 The pending element, through the index value i Get the number D i , use the number to query E Get the time when the element was added to the pending target list e , compared with the current time, if it exceeds the waiting time (such as two minutes), update A i =3 That is rejection.
[0074] 8. Use the authorized target to automatically authorize the adjacent target: traverse all A i =1The pending element, through the index value i Get LT i ; Traverse all the A j =2 The authorized element, by index value j Get authorized targets LT j ,if LT i and LT j When the attribute difference (distance or distance + speed direction) is within a certain range, update A i =2 That is, authorization. A When a new authorization target is added, this step needs to be executed again to authorize the unauthorized targets within the scope of the newly added authorization target until there are no new authorization targets.
[0075] , Automatic authorization using adjacent area authorization: When an authorized activity target in a certain control area enters an adjacent authorization area, the device in the control area contacts the device in the adjacent control area that shares the adjacent authorization area with it. After receiving the notification, the device in the adjacent control area writes a ( s,e ). Traverse the authorization list of adjacent areas LZ , get each ( s, e ),if e The current time has passed, the authorization is removed from LZ Delete from; if e Less than the current time, through s Get S s of( O,P,r ) Traverse and compare unauthorized targets A j =0 and pending targets A j =1 ,if LT j exist( O,P,r ) A j Updated to 2; After completing the authorization traversal, from LZ Delete the used adjacent area authorization, that is, the adjacent area authorization can only be used once.
[0076] 10. Use mandatory authorization: traverse the mandatory authorization list LG For each element in , get each e ,ife The current time has passed, the authorization is removed from LG Delete from; if e Less than the current time, traverse A ,Will Each unauthorized A Update to 2. After completing the authorization traversal, for the used authorizations, you need to LG Delete from the mandatory authorization list LG For the writing of , see process 12.
[0077] 11. Alarm for unauthorized and rejected targets: traversal A ,for A i =0 or A i =3 The device outputs alarm information, including local alarm through speakers and lights connected to the device, or external output through other means such as BLE and WiFi, and further alarm through mobile phones or the cloud.
[0078] 12. Processing of alarm information: After obtaining the alarm information (locally, remotely, etc.), the user can generate mandatory authorization through gestures, voice, mobile phone APP, etc. e (such as 3 seconds later) and add it to the mandatory authorization list LG middle.
[0079] It should be noted that, in the above implementation process, except for process 1 and 2, processes 3 to 12 have no clear order. In the process of implementing the above process using a computer, each process 3 to 12 is in a state of repeated execution, and the process number is only for the convenience of description. i, j It means "each one", in different processes i, j There is no correlation between the values, the same as in the step i, j The differentiated use of is only to distinguish parameters belonging to different sets.
[0080] Example 4
[0081] An electronic device comprises: one or more processors; a storage device on which one or more programs are stored; when the one or more programs are executed by the one or more processors, the one or more processors implement any of the methods described in Examples 1-3.
Claims
1. A method for target authorization based on continuous radar detection and tracking, characterized in that: The steps include, S1: Acquire control area information, where the control area information at least includes control area boundary information and entry authorization area information; S2: Continuously collect point cloud information of active targets in the prevention and control area, and detect and track active targets; S3: Detect and identify unauthorized new targets in the control area. When an unauthorized new target is detected in the entry authorization area, a waiting time is given to the new target and the target authorization status is updated to pending. Traverse the distance between the pending target and each authorized target at each acquisition time during the waiting time, as well as the movement direction of the authorized target, and determine whether there is an authorized target whose movement direction points to the pending target and the distance between the authorized target and the pending target is less than or equal to the set threshold. If so, the pending target is authorized, and the target authorization status is updated to authorized; If the angle between the line connecting the positions of the authorized target and the pending target and the moving direction of the authorized target is less than or equal to a set threshold, it is determined that the moving direction of the authorized target points to the pending target.
2. The method for target authorization based on continuous radar detection and tracking as claimed in claim 1, characterized in that: There are multiple control areas, and the monitoring ranges of radar monitoring equipment deployed in two control areas overlap to form a boundary. The two control areas are adjacent to each other, and the formed boundary is an adjacent authorization area. The adjacent authorization area is an entry authorization area shared by the two control areas; multiple control areas connected by adjacent authorization areas form a joint control area; when an authorized target is tracked to move from one control area to an adjacent authorization area, the target is given authorization to be associated with the adjacent control area of the adjacent authorization area.
3. An electronic device, comprising: one or more processors; a storage device having one or more programs stored thereon; When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1-2.
4. A method for intrusion detection based on continuous radar detection and tracking, characterized in that: The steps include, S10: Acquire control area information, where the control area information at least includes control area boundary data and entry authorization area information; S20: Continuously collect point cloud information of active targets in the prevention and control area, and detect and track active targets; S30: Detect and identify unauthorized new targets, and determine whether the first appearance position of the new target is within any entry authorization area; if not, determine that the target is an intruder; if so, give the target a waiting time and mark the target as a pending target; S40: Check whether the pending target has obtained authorization through any of the following authorization methods within the waiting time: A collects the identification information of the pending target that can be used to obtain authorization, verifies and identifies the identification information, and if it matches, authorizes the pending target and marks it as an authorized target; B. traverse the distance between the pending target and each authorized target at each collection time during the waiting time and the movement direction of the authorized target, and determine whether there is an authorized target whose movement direction points to the pending target and the distance between the authorized target and the pending target is less than or equal to the set threshold. If so, the pending target is authorized and marked as an authorized target; If the angle between the line connecting the positions of the authorized target and the pending target and the movement direction of the authorized target is less than or equal to the set threshold, the movement direction of the authorized target is considered to be pointing to the pending target; if the pending target still has not been authorized after the waiting time expires, the authorization of the target is explicitly rejected and the target is judged to be an intruder; S50: When the target is tracked to leave the current control area and exceeds the set time, the information of the target is cleared and the tracking of the target is stopped.
5. The method for intrusion detection based on continuous radar detection and tracking as claimed in claim 4, characterized in that: There are multiple control areas, and the monitoring ranges of radar monitoring equipment deployed in two control areas overlap to form a boundary. The two control areas are adjacent to each other, and the boundary formed is an adjacent authorization area. The adjacent authorization area is an entry authorization area shared by the two control areas; multiple control areas connected by adjacent authorization areas form a joint control area; the authorization method of step S40 also includes C adjacent area authorization: when an authorized target is tracked to move from one control area to an adjacent authorization area, the target is given authorization to be associated with the adjacent control area of the adjacent authorization area.
6. The method for intrusion detection based on continuous radar detection and tracking as claimed in claim 5, characterized in that: The authorization method also includes D compulsory authorization: receiving a compulsory authorization instruction issued by a user and granting compulsory authorization to the target.
7. The method for intrusion detection based on continuous radar detection and tracking as claimed in claim 4, characterized in that: Step S50 specifically includes: traversing the point cloud information of each target that was collected the last time, and determining whether the difference between the collection time and the current time exceeds a set threshold; If exceeded, determine whether the location of the target of this collection is within the entry authorization area or within the set threshold range near the boundary of the control area. If so, determine that the target has left the current control area and exceeded the set time, clear the target's information and stop tracking the target.
8. The method for intrusion detection based on continuous radar detection and tracking as claimed in claim 4, characterized in that: When a target is judged as an intruder, an alarm message is output.
9. The method for intrusion detection based on continuous radar detection and tracking as claimed in claim 4, characterized in that: The identification information includes the target's biometric information, the electronic identification carried by the target, and the electronic identification generated by the smart door lock when the target uses a legitimate key to open the door.
10. An electronic device, comprising: one or more processors; a storage device having one or more programs stored thereon; When the one or more programs are executed by the one or more processors, the one or more processors implement the method described in any one of claims 4 to 9.
Citation Information
Cited By
Intrusion detection system and method
EP4773113A1