Law enforcement action assessment and evaluation method and system based on virtual scene

Abstract

The invention relates to a law enforcement action assessment and evaluation method and system based on a virtual scene, and the method comprises the steps: obtaining the fingerprint information of a subject through VR glasses, verifying the identity of the subject, collecting the joint points of human bones after verification succeeds, determining the coordinate system of each node in the joint points, obtaining VR assessment scene data, and carrying out the assessment and evaluation of a law enforcement action. Determining a VR scene plane coordinate system with position attributes in the VR assessment scene data; establishing an index relationship between a coordinate system of each node in the joint points and a VR scene plane coordinate system with position attributes in the VR assessment scene data and a VR scene sphere coordinate system of a scene sphere in the VR glasses, so that the VR glasses enter a VR panoramic mode to perform law enforcement action assessment; and in the assessment process, according to the joint points of the human skeleton, obtaining an instantaneous law enforcement action of the subject, taking the shortest path between each joint point of a preset standard action and the joint points of the human skeleton as an evaluation standard for evaluation, and obtaining an evaluation result.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to a method and system for assessing and evaluating law enforcement actions based on a virtual scenario. Background Art

[0002] Virtual reality (VR) technology is a computer simulation system that can generate a simulated environment and provide subjects with an immersive application experience through three-dimensional dynamic vision, entity behavior interaction, etc.

[0003] With the continuous improvement of computer graphics processing technology and hardware performance, virtual reality (VR) technology has developed rapidly. VR technology simulates real or fictional environments, allowing subjects to experience various law enforcement scenarios in an immersive way.

[0004] Existing assessment technology solutions are often limited by factors such as time, space, and resources, and are unable to meet the needs of modern law enforcement assessments. They often rely on paper-based test papers or simple computer tests, making it difficult to fully assess the actual abilities and skill levels of the subjects. Summary of the invention

[0005] In order to solve the problems existing in the above-mentioned prior art, the purpose of the present invention is to propose a law enforcement action assessment and evaluation method and system based on virtual scenes. Through the virtual scenes, subjects can be assessed at any time and any place, and the practicality and pertinence of learning can be enhanced by simulating real scenes.

[0006] To achieve the above object, the present invention provides the following solutions:

[0007] A method for assessing and evaluating law enforcement actions based on a virtual scene, comprising:

[0008] Using VR glasses to obtain fingerprint information of the subject, verify the identity of the subject according to the fingerprint information, and after successful verification, collect the joint points of the human skeleton, determine the coordinate system of each node in the joint points, obtain VR assessment scene data, and determine the VR scene plane coordinate system with position attributes in the VR assessment scene data;

[0009] An index relationship is established between the coordinate system of each node in the joint point and the VR scene plane coordinate system with position attributes in the VR assessment scene data and the VR scene sphere coordinate system of the scene sphere in the VR glasses, so that the VR glasses enter the VR panoramic mode to perform law enforcement action assessment;

[0010] During the assessment process, the instantaneous law enforcement movements of the subject are obtained based on the joints of the human skeleton, and the shortest path between each joint of the preset standard movement and the joints of the human skeleton is used as an evaluation criterion. The instantaneous law enforcement movements of the subject are compared with the preset standard movements, and evaluated to obtain evaluation results.

[0011] Optionally, verifying the identity of the subject according to the fingerprint information includes:

[0012] The fingerprint information of the subject is obtained by using VR glasses; wherein the VR glasses at least include: a fingerprint collection device arranged on the outer shell of the VR glasses, and a movable spring device arranged at the lens placement position of the VR glasses;

[0013] The fingerprint information of the subject is compared with the fingerprint information in the preset personal information to verify the identity of the subject. If the subject passes the identity verification, it indicates that the verification is successful; if the subject fails the identity verification, the mobile spring device is controlled to pop out the spring to cover the lenses of the VR glasses.

[0014] Optionally, determining the coordinate system of each node in the joint point includes:

[0015] Collecting motion capture data of various points of the subject, and determining the corresponding joint points of the human skeleton according to the motion capture data;

[0016] The origin of the VR scene sphere coordinate system of the scene sphere in the VR glasses is used as the origin of the coordinate system of each node to determine the coordinate system of each node in the joint point.

[0017] Optionally, the VR assessment scene data includes: panoramic pictures, panoramic videos, panoramic animations and audio.

[0018] Optionally, establishing an index relationship between a VR scene plane coordinate system having a position attribute in the VR assessment scene data and a VR scene sphere coordinate system of a scene sphere in the VR glasses includes:

[0019] According to the division method of longitude and latitude, an index relationship is established between the VR scene plane coordinate system with position attributes in the VR assessment scene data and the VR scene sphere coordinate system of the scene sphere in the VR glasses.

[0020] Optionally, causing the VR glasses to enter a VR panoramic mode for law enforcement action assessment includes:

[0021] Acquiring vision data of the subject, and adjusting the focal length of the VR glasses according to the vision data of the subject;

[0022] Acquire the texture coordinates of the VR assessment scene data, and render the VR assessment scene data onto the VR glasses based on the texture coordinates and the established index relationship;

[0023] The vision data of the subject is obtained, and the focal length of the VR glasses is adjusted according to the vision data of the subject in combination with the image rendered on the screen, so as to use the VR glasses to conduct law enforcement action assessment.

[0024] Optionally, adjusting the focal length of the VR glasses includes:

[0025] Obtaining vision data of the subject, and adjusting the distance between the center points of the two screens of the VR glasses to be the same as the eye distance of the subject;

[0026] According to the vision data of the subject and the image rendered on the screen, the focal length of the VR glasses is adjusted so that the VR assessment scene can be clearly imaged on the retina of the subject.

[0027] Optionally, obtaining the instantaneous law enforcement action of the subject includes:

[0028] During the assessment, the acceleration data of the subject performing relevant law enforcement actions is obtained according to the joint points of the human skeleton;

[0029] Eliminate missing values ​​in the acceleration data, extract action frames, set an observation time window, preset instantaneous thresholds to filter the extracted action frames, determine the occurrence time of the law enforcement action, and obtain acceleration frame data based on the filtered action frames combined with the occurrence time of the law enforcement action;

[0030] The acceleration frame data are marked and the instantaneous law enforcement actions of the subject in the data are selected.

[0031] To achieve the above-mentioned purpose, the present invention provides a law enforcement action assessment and evaluation system based on a virtual scene, comprising:

[0032] A position acquisition module is used to obtain fingerprint information of the subject using VR glasses, verify the identity of the subject according to the fingerprint information, collect joint points of the human skeleton after successful verification, determine the coordinate system of each node in the joint points, obtain VR assessment scene data, and determine the VR scene plane coordinate system with position attributes in the VR assessment scene data;

[0033] A position calibration module is used to establish an index relationship between the coordinate system of each node in the joint point and the VR scene plane coordinate system with position attributes in the VR assessment scene data and the VR scene sphere coordinate system of the scene sphere in the VR glasses, so that the VR glasses enter the VR panoramic mode to perform law enforcement action assessment;

[0034] The assessment and evaluation module is used to obtain the subject's instantaneous law enforcement actions according to the joints of the human skeleton during the assessment process, and use the shortest path between each joint of the preset standard action and the joints of the human skeleton as the evaluation standard, compare the subject's instantaneous law enforcement actions with the preset standard actions, and evaluate and obtain the evaluation results.

[0035] The beneficial effects of the present invention are:

[0036] The present invention simulates actual law enforcement assessment scenarios by constructing a realistic virtual environment, thereby improving the accuracy and effectiveness of the assessment.

[0037] The virtual scene-based law enforcement action assessment method of the present invention can break the limitation of the difficulty in comprehensively assessing the actual ability and skill level of the subjects, and provide a flexible, convenient and efficient law enforcement assessment method. Through the virtual scene, the subjects can conduct law enforcement assessment at any time and any place, and at the same time, the practicality and pertinence of learning can be enhanced by simulating law enforcement scenes.

[0038] The present invention can conduct a comprehensive assessment and evaluation of the subjects by simulating real law enforcement assessment scenarios. This assessment method can not only examine the theoretical knowledge of the subjects, but also evaluate their practical operational capabilities and ability to deal with complex situations, thereby more accurately reflecting the real level of the subjects. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0040] Figure 1 The present invention is a flowchart of a method for assessing and evaluating law enforcement actions based on a virtual scenario according to an embodiment of the present invention. DETAILED DESCRIPTION

[0041] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0042] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0043] like Figure 1 As shown, the present embodiment discloses a method for assessing and evaluating law enforcement actions based on a virtual scene, including: using VR glasses to obtain fingerprint information of a subject, verifying the identity of the subject according to the fingerprint information, collecting joint points of the human skeleton after successful verification, determining the coordinate system of each node in the joint points, obtaining VR assessment scene data, and determining the VR scene plane coordinate system with position attributes in the VR assessment scene data; the VR assessment scene data includes: panoramic pictures, panoramic videos, panoramic animations, and audio; establishing an index relationship between the coordinate system of each node in the joint points and the VR scene plane coordinate system with position attributes in the VR assessment scene data and the VR scene sphere coordinate system of the scene sphere in the VR glasses, so that the VR glasses enter the VR panoramic mode for law enforcement action assessment; during the assessment process, according to the joint points of the human skeleton, the instantaneous law enforcement action of the subject is obtained, and the shortest path between each joint point of the preset standard action and the joint point of the human skeleton is used as the evaluation standard, and the instantaneous law enforcement action of the subject is compared with the preset standard action, and evaluated to obtain the evaluation result.

[0044] Verifying the identity of the subject according to the fingerprint information includes: obtaining the fingerprint information of the subject by using VR glasses; wherein the VR glasses at least include: a fingerprint collection device arranged on the outer shell of the VR glasses, and a mobile spring device arranged at the lens placement position of the VR glasses; verifying the identity of the subject according to the fingerprint information of the subject and the fingerprint information in the preset personal information, if the subject passes the identity verification, it indicates that the verification is successful; if the subject fails the identity verification, the mobile spring device is controlled to eject the spring to cover the lens of the VR glasses, specifically:

[0045] Compare the fingerprint information with the fingerprint information in the preset personal information to see if they are consistent; if the fingerprint information is consistent with the fingerprint information in the preset personal information, determine that the subject has passed the identity authentication; if the fingerprint information is inconsistent with the fingerprint information in the preset personal information, determine whether the number of comparisons is greater than the number threshold; if the number of comparisons is not greater than the number threshold, return to execute the step of collecting the subject's fingerprint information through the VR glasses; if the number of comparisons is greater than the number threshold, determine that the subject has not passed the identity authentication, and control the mobile shrapnel device to pop out the shrapnel to cover the lenses of the VR glasses.

[0046] Determining the coordinate system of each node in the joint point includes: collecting motion capture data of each point of the subject, and determining the corresponding joint point of the human skeleton based on the motion capture data; using the origin of the VR scene sphere coordinate system of the scene sphere in the VR glasses as the origin of the coordinate system of each node, to determine the coordinate system of each node in the joint point.

[0047] Establishing an index relationship between a VR scene plane coordinate system with position attributes in VR assessment scene data and a VR scene spherical coordinate system of a scene ball in VR glasses includes: establishing an index relationship between a VR scene plane coordinate system with position attributes in VR assessment scene data and a VR scene spherical coordinate system of a scene ball in VR glasses according to the division method of longitude and latitude.

[0048] Making the VR glasses enter the VR panoramic mode for law enforcement action assessment includes: obtaining the subject's vision data, and adjusting the focal length of the VR glasses according to the subject's vision data; obtaining the texture coordinates of the VR assessment scene data, and rendering the VR assessment scene data onto the VR glasses based on the texture coordinates and the established index relationship; obtaining the subject's vision data, and adjusting the focal length of the VR glasses according to the subject's vision data and the image rendered on the screen, so as to use the VR glasses to conduct law enforcement action assessment.

[0049] Obtaining the texture coordinates of the VR assessment scene data, and rendering the VR assessment scene data onto the VR glasses based on the texture coordinates and the established index relationship includes: obtaining the texture coordinates of the VR assessment scene data; loading the vertex shader and the fragment shader, and compiling the vertex shader and the fragment shader; creating a rendering object, binding the compiled vertex shader and the compiled fragment shader to the rendering object, and compiling the rendering object to obtain the compiled program, and obtaining the texture coordinate handle in the rendering object; based on the texture coordinates, the index relationship and the texture coordinate handle, using the compiled program, rendering the real-time panoramic video stream onto the model on the VR glasses.

[0050] During the law enforcement action assessment, the VR main perspective changes with the deflection of the VR glasses' gyroscope. When the picture with position attributes is played to the corresponding playback window, the playback window follows the movement of the VR glasses' gyroscope; when the VR glasses' gyroscope rotates along the Z-axis, the VR main perspective will have a Y-axis deflection angle, and the playback window corresponding to the picture will also rotate along the Z-axis, causing the Y-axis to have the same deflection angle; when the VR glasses' gyroscope rotates along the Y-axis, the VR main perspective will have a Z-axis deflection angle, and the playback window corresponding to the picture will also rotate along the Y-axis, causing the Z-axis to have the same deflection angle; when the VR glasses' gyroscope rotates along the X-axis, the VR main perspective will have a Z-axis deflection angle, and the playback window corresponding to the picture will also rotate along the X-axis, causing the Z-axis to have the same deflection angle.

[0051] Adjusting the focal length of the VR glasses includes: obtaining the subject's vision data, obtaining the subject's vision data, and adjusting the distance between the center points of the two screens of the VR glasses to be the same as the subject's eye distance; adjusting the focal length of the VR glasses based on the subject's vision data in combination with the image rendered on the screen so that the VR assessment scene is clearly imaged on the subject's retina; wherein each change in focal length takes 0.2-2s, and the focal length stabilization time is 5-20s in each stage.

[0052] Acquiring the instantaneous law enforcement actions of the subject includes: during the assessment process, obtaining the acceleration data of the subject making relevant law enforcement actions according to the joint points of the human skeleton; eliminating the missing values ​​in the acceleration data, extracting the action frames, setting the observation time window, and filtering the extracted action frames with a preset instantaneous threshold to determine the occurrence time of the law enforcement action, and obtaining the acceleration frame data according to the filtered action frames and the occurrence time of the law enforcement action; marking the acceleration frame data and selecting the instantaneous law enforcement actions of the subject in the data.

[0053] Obtaining the preset standard action includes: for the acceleration signal collected from the instructor during the law enforcement assessment, the action frame is extracted to generate a sample segment, and the instantaneous firing action signal is selected to constitute the instructor's law enforcement assessment data, and the high-frequency signal that lasts within 3ms after the occurrence time of the action signal is intercepted. Then, the DTW algorithm is used to select the best frame data during multiple law enforcement assessments of the instructor as the instantaneous standardization evaluation index of the law enforcement assessment.

[0054] The present embodiment also discloses a law enforcement action assessment and evaluation system based on a virtual scene, including: a position acquisition module, used to use VR glasses to obtain the fingerprint information of the subject, verify the identity of the subject according to the fingerprint information, and after successful verification, collect the joints of the human skeleton, determine the coordinate system of each node in the joints, obtain VR assessment scene data, and determine the VR scene plane coordinate system with position attributes in the VR assessment scene data; a position calibration module, used to establish an index relationship between the coordinate system of each node in the joints and the VR scene plane coordinate system with position attributes in the VR assessment scene data and the VR scene sphere coordinate system of the scene sphere in the VR glasses, so that the VR glasses enter the VR panoramic mode to perform law enforcement action assessment; an assessment and evaluation module, used to obtain the instantaneous law enforcement action of the subject according to the joints of the human skeleton during the assessment process, use the shortest path between each joint of the preset standard action and the joints of the human skeleton as the evaluation standard, compare the instantaneous law enforcement action of the subject with the preset standard action, and evaluate and obtain the evaluation result.

[0055] The embodiments described above are only descriptions of the preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Without departing from the design spirit of the present invention, various modifications and improvements made to the technical solutions of the present invention by ordinary technicians in this field should all fall within the protection scope determined by the claims of the present invention.

Claims

1. A method for assessing law enforcement actions based on virtual scenes, characterized in that: include: Using VR glasses to obtain fingerprint information of the subject, verify the identity of the subject according to the fingerprint information, and after successful verification, collect the joint points of the human skeleton, determine the coordinate system of each node in the joint points, obtain VR assessment scene data, and determine the VR scene plane coordinate system with position attributes in the VR assessment scene data; An index relationship is established between the coordinate system of each node in the joint point and the VR scene plane coordinate system with position attributes in the VR assessment scene data and the VR scene sphere coordinate system of the scene sphere in the VR glasses, so that the VR glasses enter the VR panoramic mode to perform law enforcement action assessment; During the assessment process, the instantaneous law enforcement movements of the subject are obtained based on the joints of the human skeleton, and the shortest path between each joint of the preset standard movement and the joints of the human skeleton is used as an evaluation criterion. The instantaneous law enforcement movements of the subject are compared with the preset standard movements, and evaluated to obtain evaluation results.

2. The virtual scene-based law enforcement action assessment and evaluation method according to claim 1 is characterized in that: Verifying the identity of the subject based on the fingerprint information includes: The fingerprint information of the subject is obtained by using VR glasses; wherein the VR glasses at least include: a fingerprint collection device arranged on the outer shell of the VR glasses, and a movable spring device arranged at the lens placement position of the VR glasses; The fingerprint information of the subject is compared with the fingerprint information in the preset personal information to verify the identity of the subject. If the subject passes the identity verification, it indicates that the verification is successful; if the subject fails the identity verification, the mobile spring device is controlled to pop out the spring to cover the lenses of the VR glasses.

3. The virtual scene-based law enforcement action assessment and evaluation method according to claim 1 is characterized in that: Determining the coordinate system of each node in the joint point includes: Collecting motion capture data of various points of the subject, and determining the corresponding joint points of the human skeleton according to the motion capture data; The coordinate system of each node in the joint point is determined by taking the origin of the VR scene sphere coordinate system of the scene sphere in the VR glasses as the origin of the coordinate system of each node.

4. The virtual scene-based law enforcement action assessment and evaluation method according to claim 1 is characterized in that: The VR assessment scene data includes: panoramic pictures, panoramic videos, panoramic animations and audio.

5. The virtual scene-based law enforcement action assessment and evaluation method according to claim 1 is characterized in that: Establishing an index relationship between the VR scene plane coordinate system with position attributes in the VR assessment scene data and the VR scene sphere coordinate system of the scene sphere in the VR glasses includes: According to the division method of longitude and latitude, an index relationship is established between the VR scene plane coordinate system with position attributes in the VR assessment scene data and the VR scene sphere coordinate system of the scene sphere in the VR glasses.

6. The virtual scene-based law enforcement action assessment and evaluation method according to claim 1 is characterized in that: The VR glasses are made to enter the VR panoramic mode to conduct law enforcement action assessment, including: Acquiring vision data of the subject, and adjusting the focal length of the VR glasses according to the vision data of the subject; Acquire the texture coordinates of the VR assessment scene data, and render the VR assessment scene data onto the VR glasses based on the texture coordinates and the established index relationship; The vision data of the subject is obtained, and the focal length of the VR glasses is adjusted according to the vision data of the subject in combination with the image rendered on the screen, so as to use the VR glasses to conduct law enforcement action assessment.

7. The virtual scene-based law enforcement action assessment and evaluation method according to claim 6 is characterized in that: Adjusting the focal length of the VR glasses includes: Obtaining vision data of the subject, and adjusting the distance between the center points of the two screens of the VR glasses to be the same as the eye distance of the subject; According to the vision data of the subject and the image rendered on the screen, the focal length of the VR glasses is adjusted so that the VR assessment scene can be clearly imaged on the retina of the subject.

8. The virtual scene-based law enforcement action assessment and evaluation method according to claim 1 is characterized in that: Acquiring the instantaneous law enforcement actions of the subject includes: During the assessment, the acceleration data of the subject performing relevant law enforcement actions is obtained according to the joint points of the human skeleton; Eliminate missing values ​​in the acceleration data, extract action frames, set an observation time window, preset instantaneous thresholds to filter the extracted action frames, determine the occurrence time of the law enforcement action, and obtain acceleration frame data based on the filtered action frames combined with the occurrence time of the law enforcement action; The acceleration frame data are marked and the instantaneous law enforcement actions of the subject in the data are selected.

9. A system for evaluating and assessing law enforcement actions based on virtual scenes as described in any one of claims 1 to 8, characterized in that: include: A position acquisition module is used to obtain fingerprint information of the subject using VR glasses, verify the identity of the subject according to the fingerprint information, collect joint points of the human skeleton after successful verification, determine the coordinate system of each node in the joint points, obtain VR assessment scene data, and determine the VR scene plane coordinate system with position attributes in the VR assessment scene data; A position calibration module is used to establish an index relationship between the coordinate system of each node in the joint point and the VR scene plane coordinate system with position attributes in the VR assessment scene data and the VR scene sphere coordinate system of the scene sphere in the VR glasses, so that the VR glasses enter the VR panoramic mode to perform law enforcement action assessment; The assessment and evaluation module is used to obtain the subject's instantaneous law enforcement actions according to the joints of the human skeleton during the assessment process, and use the shortest path between each joint of the preset standard action and the joints of the human skeleton as the evaluation standard, compare the subject's instantaneous law enforcement actions with the preset standard actions, and evaluate and obtain the evaluation results.

Images