Data security sharing method and system for personnel video image and positioning

Abstract

The invention relates to the technical field of data security sharing methods and systems, in particular to a data security sharing method and system for personnel video image and positioning. Comprising a video image acquisition unit, a data preprocessing unit, an intelligent analysis and feature extraction unit, a dynamic three-dimensional reconstruction and positioning unit, a secure storage and privacy calculation unit, a shared service scheduling unit and a data analysis and mining unit. The video image acquisition unit is deployed in a monitoring area, acquires video images and positioning data of personnel in real time through carrying a high-definition camera on high-altitude acquisition equipment and carrying GPS (Global Positioning System) positioning equipment by the personnel, and then transmits the acquired original data to the preprocessing unit; the data preprocessing unit is used for carrying out cleaning, denoising and format conversion preprocessing operation on the collected original video image and positioning data; according to the method, efficient, safe and intelligent personnel video image and positioning data sharing is realized, and powerful technical support is provided for safety management, decision support and multi-party cooperation.

Description

Technical Field

[0001] The present invention relates to the technical field of data security sharing methods and systems, and particularly to a data security sharing method and system for personnel video images and positioning. Background Art

[0002] With the rapid development of technology, video surveillance and positioning technologies have become an indispensable part of modern society. In various places, such as public places, enterprise parks, residential communities, etc., surveillance cameras and positioning devices are widely deployed to ensure personnel safety, monitor important areas, and manage the flow of people.

[0003] For example, in the patent with the prior art publication number CN109814068B, the invention discloses a construction site personnel positioning system, including a first positioning module for obtaining data identifiers of personnel to be positioned, an image acquisition module for collecting video images within the construction site, a second positioning module for obtaining second position information of the personnel to be positioned according to the video images, a control center for analyzing according to the data identifiers and second position information of the personnel to be positioned to judge the positioning situation of the personnel to be positioned and generating alarm information according to the positioning situation, a display module for respectively displaying a first positioning mark and a second positioning mark of the personnel to be positioned on a map according to the data identifiers and second position information of the personnel to be positioned, an alarm module for issuing an alarm according to the alarm information, and a communication module for communication between each module.

[0004] It is found in the use of this system that the image acquisition module in this system needs to set multiple cameras in different working areas, which increases the cost and is not convenient for generating a dynamic heat map, reducing the use effect. Summary of the Invention

[0005] To solve the above technical problems, the present invention provides a data security sharing method and system for personnel video images and positioning, which realizes efficient, safe, and intelligent sharing of personnel video images and positioning data, and provides strong technical support for security management, decision-making support, and multi-party collaboration.

[0006] A data security sharing system for personnel video images and positioning of the present invention includes a video image acquisition unit, a data preprocessing unit, an intelligent analysis and feature extraction unit, a dynamic three-dimensional reconstruction and positioning unit, a secure storage and privacy computing unit, a shared service scheduling unit, and a data analysis and mining unit;

[0007] The video image acquisition unit is deployed in the monitoring area, and through high-altitude acquisition equipment carrying high-definition cameras and GPS positioning devices carried by personnel, it collects personnel video images and positioning data in real time, and then transmits the collected original data to the preprocessing unit;

[0008] The data preprocessing unit performs preprocessing operations such as cleaning, denoising, and format conversion on the collected original video images and positioning data, and ensures data quality through a combination of automated algorithms and manual review;

[0009] The intelligent analysis and feature extraction unit receives the data from the data preprocessing unit, integrates a multi-modal feature extraction engine, constructs an abnormal behavior detection model based on deep learning, and outputs spatio-temporal feature vectors and behavior labels, which are used to provide behavior trajectory features to the 3D reconstruction unit and submit feature indexes to the secure storage unit;

[0010] The dynamic 3D reconstruction and positioning unit receives the feature data from the intelligent analysis unit, uses SLAM technology to realize 3D scene modeling, combines UWB / Bluetooth beacons to realize passive positioning, and based on the particle filter multi-target tracking algorithm, generates a real-time personnel movement heat map, and provides spatio-temporal trajectory data packets for the shared service unit;

[0011] The secure storage and privacy computing unit is used to provide basic storage services for all units. By constructing a secure safe, it realizes sharded encrypted storage of data, deploys a federated learning middleware, and supports feature comparison in multi-party secure computing scenarios;

[0012] The shared service scheduling unit is used to construct a RESTful API service bus, supports the on-demand combination of video streams, positioning data, and analysis report product forms. By developing a QoS quality assurance engine, it realizes adaptive adjustment of bandwidth from 64 Kbps to 8 Mbps, which helps to ensure that users can obtain a stable and smooth service experience in different network environments. Through API interfaces and data exchange protocols, it realizes real-time sharing and exchange of data;

[0013] The data analysis and mining unit deeply analyzes and mines the stored data, extracts valuable information, applies algorithms such as machine learning and deep learning for data analysis, and generates visual reports and warning information; through high-definition cameras carried by mobile devices and GPS positioning devices carried by personnel, video images and positioning data are collected in real time to ensure the timeliness and comprehensiveness of the data. By integrating a multi-modal feature extraction engine and a deep learning model, it can accurately detect abnormal behaviors, output spatio-temporal feature vectors and behavior labels, provide high-quality input for subsequent analysis, realize efficient, secure, and intelligent sharing of personnel video images and positioning data, and provide strong technical support for security management, decision-making support, and multi-party collaboration.

[0014] Preferably, the shared service scheduling unit further includes a service orchestration engine unit for combining and orchestrating multiple services in the shared service scheduling unit according to specific logics and processes. Through the service orchestration engine, this unit can combine and encapsulate different service units according to business requirements to form a complete business process and service scenario.

[0015] Preferably, in the data preprocessing unit, it consists of the following steps through a combination of automated algorithms and manual review:

[0016] 1. Automated algorithms:

[0017] 1.1. Use methods such as interpolation, mean filling, and deletion of missing values to process missing items in the data, apply statistical methods and machine learning algorithms to identify and process outliers, and remove duplicate data through database indexing methods, thereby cleaning the data;

[0018] 1.2. Apply filtering techniques to remove noise in the images, and use smoothing algorithms to reduce jitter and errors in the positioning data;

[0019] 1.3. Use the FFmpeg tool to convert videos into a unified MP4 format, and convert images into a unified JPEG format;

[0020] 1.4. Set quality indicators for image clarity and positioning accuracy, and use automated algorithms for preliminary evaluation. According to the quality indicators, classify the data into three categories: high quality, medium quality, and low quality, and make corresponding markings;

[0021] 2. Manual review:

[0022] 2.1. Assign the data marked as low quality or questionable by the automated algorithm to the reviewers for manual review, including image clarity and positioning accuracy;

[0023] 2.2. Use specialized review tools or interfaces to facilitate reviewers to view, process, and mark the data, and allow reviewers to provide feedback on the results of the automated algorithm in order to continuously optimize the algorithm performance;

[0024] 2.3. Provide professional training to the reviewers to ensure familiarity with the review standards and processes, and at the same time regularly review the review results to ensure the consistency of the review quality;

[0025] 2.4. Continuously optimize the automated algorithm based on the feedback and review results of the reviewers to improve the accuracy and efficiency of data preprocessing; through continuous iteration of the automated algorithm and manual review process, gradually improve the data quality, ensure the collaborative work between the automated algorithm and manual review, and form a closed-loop feedback mechanism.

[0026] Preferably, the video image acquisition unit further includes a jitter compensation system for compensating for the jitter of the high-definition camera carried by the high-altitude acquisition device, enhancing the effect of the high-definition camera to stably acquire video images when shooting in the monitoring area;

[0027] The jitter compensation system includes a jitter detection unit, a jitter compensation unit, an image stabilization unit and a data transmission unit;

[0028] The jitter detection unit uses a jitter detection algorithm to analyze IMU data, identify the direction and amplitude of the jitter, and combine it with GPS data to determine whether the jitter is caused by external factors;

[0029] The jitter compensation unit uses a compensation control algorithm to generate a three-axis gimbal adjustment command based on the jitter detection results, and adjusts the camera angle in real time to offset the jitter;

[0030] The image stabilization unit post-processes the video frames and uses an image stabilization algorithm to eliminate residual jitter;

[0031] The data transmission unit transmits the stable video images and positioning data to the preprocessing unit; the jitter compensation system can effectively improve the stability and clarity of the video image acquisition unit, ensuring the accuracy and reliability of the monitoring data.

[0032] A data security sharing method for personnel video images and positioning includes the following steps:

[0033] S1. Deploy high-altitude data collection equipment in the monitoring area, equipped with high-definition cameras to collect real-time video images of personnel. Personnel carry GPS positioning equipment with them to collect positioning data in real time;

[0034] S2, cleaning the original video image, removing blur, jitter and other low-quality frames, denoising the positioning data, and removing outliers;

[0035] S3, integrated multimodal feature extraction engine, extracts spatiotemporal features from video images and positioning data, builds an abnormal behavior detection model based on deep learning, outputs spatiotemporal feature vectors and behavior labels, sends behavior trajectory features to the 3D reconstruction unit, and submits feature indexes to the secure storage unit;

[0036] S4, using SLAM technology to achieve scene 3D modeling, generate high-precision 3D scenes, combined with UWB / Bluetooth beacons, achieve non-sensing positioning, accurately track personnel positions, and use particle filter-based multi-target tracking algorithms to generate personnel movement heat maps in real time;

[0037] S5. Build a safe, encrypt and store data in shards to ensure data security, deploy federated learning middleware, support feature comparison in multi-party secure computing scenarios, and protect data privacy;

[0038] S6. Build a RESTful API service bus to support on-demand combination of video streams, location data, analysis reports and other product forms, and realize real-time sharing and exchange of data through API interfaces and data exchange protocols;

[0039] S7. Deeply analyze and mine the stored data, extract valuable information, apply algorithms such as machine learning and deep learning for data analysis, and generate visual reports and warning information;

[0040] S8. Encrypt the transmitted and stored data and implement strict access control;

[0041] S9. Monitor the system operation status in real time, detect and handle anomalies in a timely manner.

[0042] Preferably, the high-altitude acquisition device includes a control device, a housing, a support plate, multiple groups of propeller power components, multiple groups of three-axis pan-tilt heads, multiple groups of cameras, and an elastic balloon. The support plate is installed on the outer wall of the housing. Multiple groups of propeller power components are all installed on the outer wall of the support plate. Multiple groups of three-axis pan-tilt heads are all installed at the bottom of the outer wall of the support plate. Multiple groups of cameras are respectively installed on the mobile ends of multiple groups of three-axis pan-tilt heads. The elastic balloon is installed inside the housing, and an air vent is provided at the lower part of the outer wall of the housing. The control device is connected to the bottom end of the housing. The control device is used to control the lifting and moving of the housing, and the control device protects multiple groups of cameras by cooperating with the support plate; by filling hydrogen into the elastic balloon through the control device, after the elastic balloon expands, the air inside the housing is squeezed out, so that the housing generates buoyancy and moves upward. The housing drives multiple groups of cameras to rise to a high altitude, so that multiple groups of cameras can take high-altitude images of personnel's image information, reducing the occlusion of obstacles and at the same time reducing the number of cameras used. By using hydrogen as the rising power, the energy consumption of the device is reduced and the in-air shooting time is increased. The position of the housing is corrected by multiple groups of propeller power components. By setting multiple groups of three-axis pan-tilt heads, the shooting stability of the cameras is improved. By pumping out the hydrogen in the elastic balloon, it is convenient to store the housing downward. After the housing moves downward, multiple groups of cameras are protected by the cooperation of the support plate and the control device, thereby reducing the damage of multiple groups of cameras.

[0043] Preferably, the control device includes a power device, a chassis, a base, a housing, a delivery cylinder, an air pipe, a cable, and a sealing ring. The base is provided at the bottom end of the chassis, the housing is installed at the top end of the chassis, the top end of the delivery cylinder is communicated with the housing, the bottom end of the delivery cylinder extends outside the chassis, the top ends of the air pipe and the cable are both connected to the cover body, and the air pipe is communicated with the inside of the elastic balloon. The sealing ring is installed on the outer side wall of the housing. A power device is arranged inside the chassis. The bottom ends of the air pipe and the cable are connected to the power device, and the air pipe and the cable pass through the inside of the delivery cylinder. The power device is used for controlling the winding and unwinding of the air pipe and the cable, and the power device is used for providing electricity and hydrogen. By unwinding the air pipe and the cable through the power device, while the power device supplies power to the cable, hydrogen is conveyed into the air pipe, and the air pipe conveys the hydrogen into the elastic balloon, so that the cover body rises. The cable supplies power to multiple groups of propeller power components, multiple groups of three-axis pan-tilt heads, and multiple groups of cameras, thereby improving the convenience of power supply for image shooting. When it rains outdoors, the rainwater enters the delivery cylinder through the housing, and the rainwater is guided downward and discharged through the delivery cylinder, thereby avoiding the accumulation of water inside the housing. After the cover body descends, the support plate is closed with the housing. At this time, multiple groups of three-axis pan-tilt heads and multiple groups of cameras are received inside the housing, thereby improving its protection effect.

[0044] Preferably, the power device includes a supply device, a driving device, a cable shaft, an electromagnetic slip ring, a delivery cylinder, a delivery box, and two groups of guide wheels. The cable shaft is rotatably installed on the inner side wall of the support plate, the electromagnetic slip ring is arranged between the support plate and the cable shaft, the delivery cylinder is rotatably installed on the inner side wall of the support plate, the delivery box is installed on the inner side wall of the support plate, the delivery cylinder is rotatably connected to the delivery box, and the delivery cylinder is communicated with the inside of the delivery box. The delivery box is communicated with the supply device. The supply device is used for conveying hydrogen into the delivery box. A driving device is arranged inside the support plate. The driving device is used for driving the cable shaft and the delivery cylinder to rotate. The two groups of guide wheels are both installed on the inner side wall of the delivery cylinder. The bottom end of the cable is connected to the cable shaft, and the bottom end of the air pipe is communicated with the delivery cylinder. The support plate supplies power to the cable shaft through the electromagnetic slip ring, so that the cable shaft supplies power to the cable. The supply device supplies hydrogen into the delivery cylinder through the delivery box, so that the delivery cylinder supplies hydrogen to the air pipe. By driving the cable shaft and the delivery cylinder to rotate through the driving device, the cable shaft and the delivery cylinder wind or unwind the air pipe and the cable at the same time, so that the cover body drives multiple groups of cameras to move up and down for adjustment. The two groups of guide wheels guide and support the air pipe and the cable, thereby avoiding abrasion between the air pipe and the cable and the delivery cylinder and improving the protection effect.

[0045] Preferably, the driving device includes a support seat, a worm, a motor and two sets of worm gears, the support seat is installed on the inner wall of the cover body, the worm is rotatably installed on the support seat, the motor is installed on the outer wall of the support seat, the motor output end is connected to the worm, the two sets of worm gears are respectively installed on the cable shaft and the conveying drum, and the two sets of worm gears are meshed with the worm; the worm is driven to rotate by the motor, so that the worm drives the cable shaft and the conveying drum to rotate respectively through the meshing with the two sets of worm gears, thereby improving the synchronization of the movement of the air pipe and the cable, and at the same time, due to the meshing of the worm wheel and the worm, the positioning effect of the cable shaft and the conveying drum is improved.

[0046] Preferably, the supply device includes a storage tank, a pump body and a valve, the storage tank and the pump body are respectively mounted on the inner wall of the support plate, the pump body is connected between the storage tank and the delivery box, and the valve is connected on the delivery end of the pump body; the pump body delivers the hydrogen in the storage tank to the delivery box, thereby inflating the elastic balloon and rising, and the pump body sucks air into the delivery box, thereby discharging the hydrogen in the elastic balloon into the storage tank, thereby improving the convenience of controlling the descent of the elastic balloon.

[0047] Compared with the prior art, the beneficial effects of the present invention are as follows: through a mobile device equipped with a high-definition camera and a GPS positioning device carried by personnel, video images and positioning data are collected in real time to ensure the timeliness and comprehensiveness of the data, and a multimodal feature extraction engine and a deep learning model are integrated to accurately detect abnormal behavior, output spatiotemporal feature vectors and behavior labels, and provide high-quality input for subsequent analysis, thereby realizing efficient, safe, and intelligent sharing of personnel video images and positioning data, and providing strong technical support for security management, decision support, and multi-party collaboration. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 It is a schematic diagram of the system structure of the present invention;

[0049] Figure 2 It is a structural diagram of a data preprocessing unit and an intelligent analysis and feature extraction unit;

[0050] Figure 3 is a structural diagram of a data preprocessing unit;

[0051] Figure 4 It is a schematic diagram of the axonometric structure of the connection between the cover body and the support plate;

[0052] Figure 5 It is a schematic diagram of the axonometric structure of the connection between the mask body and the elastic balloon, etc.;

[0053] Figure 6 It is a schematic diagram of the axonometric partial structure of the connection between the support plate and the propeller power assembly, etc.;

[0054] Figure 7 It is a schematic diagram of the axonometric structure of the connection between the chassis and the shell;

[0055] Figure 8 It is an isometric structural schematic diagram of the connection between the chassis and the conveying cylinder, etc.;

[0056] Figure 9 It is an isometric partial structural schematic diagram of the connection between the support base and the worm, etc.;

[0057] Figure 10 It is an isometric partial structural schematic diagram of the connection between the conveying cylinder and the guide wheel, etc.;

[0058] Figure 11 It is an isometric partial structural schematic diagram of the connection between the storage tank and the pump body, etc.

[0059] Reference numerals in the drawings: 101, cover body; 102, support plate; 103, propeller power assembly; 104, three-axis cloud platform; 105, camera; 106, elastic balloon; 201, chassis; 202, base; 203, housing; 204, conveying cylinder; 205, air pipe; 206, cable; 207, sealing ring; 301, cable reel; 302, electromagnetic slip ring; 303, conveying cylinder; 304, conveying box; 305, guide wheel; 401, support base; 402, worm; 403, motor; 404, worm gear; 501, storage tank; 502, pump body; 503, valve. Detailed implementation manners

[0060] To facilitate the understanding of the present invention, the present invention will be described more comprehensively below with reference to the relevant drawings. The present invention can be implemented in many different forms and is not limited to the embodiments described herein. On the contrary, these embodiments are provided to make the disclosure of the present invention more thorough and comprehensive.

[0061] Embodiment 1

[0062] A data security sharing method and system for personnel video images and positioning of the present invention includes a video image acquisition unit, a data preprocessing unit, an intelligent analysis and feature extraction unit, a dynamic three-dimensional reconstruction and positioning unit, a secure storage and privacy computing unit, a sharing service scheduling unit, and a data analysis and mining unit;

[0063] The video image acquisition unit is deployed in the monitoring area. Through high-altitude acquisition equipment equipped with a high-definition camera and a GPS positioning device carried by personnel, personnel video images and positioning data are collected in real time, and then the collected original data is transmitted to the preprocessing unit;

[0064] The data preprocessing unit performs cleaning, denoising, and format conversion preprocessing operations on the collected original video images and positioning data, and ensures the data quality through a combination of automated algorithms and manual review;

[0065] The intelligent analysis and feature extraction unit receives the data from the data preprocessing unit, integrates a multi-modal feature extraction engine, constructs an abnormal behavior detection model based on deep learning, and outputs spatio-temporal feature vectors and behavior labels, which are used to provide behavior trajectory features to the 3D reconstruction unit and submit feature indexes to the secure storage unit;

[0066] The dynamic 3D reconstruction and positioning unit receives the feature data from the intelligent analysis unit, uses SLAM technology to realize 3D scene modeling, combines UWB / Bluetooth beacons to realize passive positioning, and based on the multi-target tracking algorithm of particle filter, generates a personnel movement heat map in real time, and provides spatio-temporal trajectory data packets for the shared service unit;

[0067] The secure storage and privacy computing unit is used to provide basic storage services for all units. By constructing a secure safe, it realizes sharded encrypted storage of data, deploys a federated learning middleware, and supports feature comparison in the scenario of multi-party secure computing;

[0068] The shared service scheduling unit is used to construct a RESTful API service bus, support the on-demand combination of video streams, positioning data, and analysis report product forms. By developing a QoS quality assurance engine, it realizes the adaptive adjustment of bandwidth from 64 Kbps to 8 Mbps, which helps to ensure that users can obtain a stable and smooth service experience in different network environments. Through API interfaces and data exchange protocols, it realizes the real-time sharing and exchange of data;

[0069] The data analysis and mining unit deeply analyzes and mines the stored data, extracts valuable information, applies algorithms such as machine learning and deep learning for data analysis, and generates visualization reports and warning information;

[0070] The shared service scheduling unit also includes a service orchestration engine unit: which is used to combine and orchestrate multiple services in the shared service scheduling unit according to specific logic and processes. Through the service orchestration engine, this unit can combine and encapsulate different service units according to business requirements to form a complete business process and service scenario;

[0071] In the data preprocessing unit, it is composed of the following steps through a combination of automated algorithms and manual review:

[0072] 1. Automated algorithms:

[0073] 1.1. Use methods such as interpolation, mean filling, and deletion of missing values to process missing items in the data, apply statistical methods and machine learning algorithms to identify and process outliers, and remove duplicate data through database indexing methods, thereby cleaning the data;

[0074] 1.2. Apply filtering techniques to remove noise in images, and use smoothing algorithms to reduce jitter and errors in positioning data;

[0075] 1.3. Convert the video to a unified MP4 format and the images to a unified JPEG format using the FFmpeg tool;

[0076] 1.4. Set quality indicators for image clarity and positioning accuracy, and use an automated algorithm for preliminary evaluation. According to the quality indicators, classify the data into three categories: high quality, medium quality, and low quality, and make corresponding markings;

[0077] 2. Manual review:

[0078] 2.1. Assign the data marked as low quality or questionable by the automated algorithm to reviewers for manual review, including image clarity and positioning accuracy;

[0079] 2.2. Use a dedicated review tool or interface to facilitate reviewers in viewing, processing, and marking the data, and allow reviewers to provide feedback on the results of the automated algorithm to continuously optimize the algorithm performance;

[0080] 2.3. Provide professional training to reviewers to ensure familiarity with the review standards and processes, and at the same time regularly review the review results to ensure the consistency of review quality;

[0081] 2.4. Continuously optimize the automated algorithm based on the feedback from reviewers and the review results to improve the accuracy and efficiency of data preprocessing;

[0082] The video image acquisition unit further includes a jitter compensation system for compensating for the jitter of the high-definition camera carried by the high-altitude acquisition device, enhancing the effect of the high-definition camera being able to stably acquire video images when shooting in the monitoring area;

[0083] The jitter compensation system includes a jitter detection unit, a jitter compensation unit, an image stabilization unit, and a data transmission unit;

[0084] The jitter detection unit uses a jitter detection algorithm to analyze the IMU data, identify the direction and amplitude of the jitter, and at the same time combine the GPS data to determine whether the jitter is caused by external factors;

[0085] The jitter compensation unit uses a compensation control algorithm to generate three-axis gimbal adjustment instructions according to the jitter detection results, and adjusts the camera angle in real time to offset the jitter;

[0086] The image stabilization unit performs post-processing on the video frames and uses an image stabilization algorithm to eliminate the residual jitter;

[0087] The data transmission unit transmits the stabilized video images and positioning data to the preprocessing unit;

[0088] In this embodiment, a high-definition camera is carried by a mobile device and a GPS positioning device is carried by a person to collect video images and positioning data in real time, ensuring the timeliness and comprehensiveness of the data. By integrating a multi-modal feature extraction engine and a deep learning model, abnormal behaviors can be accurately detected, and spatio-temporal feature vectors and behavior labels can be output, providing high-quality input for subsequent analysis. Efficient, secure, and intelligent sharing of personnel video images and positioning data is achieved, providing strong technical support for security management, decision-making support, and multi-party collaboration. The jitter compensation system can effectively improve the stability and clarity of the video image acquisition unit, ensuring the accuracy and reliability of the monitoring data. By continuously iterating the automated algorithm and the manual review process, the data quality is gradually improved, and the collaborative work between the automated algorithm and the manual review is ensured, forming a closed-loop feedback mechanism.

[0089] Embodiment 2

[0090] Based on Embodiment 1, a data security sharing method and system for personnel video images and positioning include the following steps:

[0091] S1. Deploy high-altitude acquisition equipment in the monitoring area, equipped with a high-definition camera to collect personnel video images in real time. A person carries a GPS positioning device to collect positioning data in real time;

[0092] S2. Clean the original video images to remove low-quality frames such as blurs and jitters, and denoise the positioning data to eliminate outliers;

[0093] S3. Integrate a multi-modal feature extraction engine to extract spatio-temporal features from the video images and positioning data, construct an abnormal behavior detection model based on deep learning, output spatio-temporal feature vectors and behavior labels, send the behavior trajectory features to the 3D reconstruction unit, and submit the feature index to the secure storage unit;

[0094] S4. Use SLAM technology to achieve 3D scene modeling, generate a high-precision 3D scene, combine with UWB / Bluetooth beacons to achieve passive positioning, accurately track the personnel's position, and based on the particle filter-based multi-target tracking algorithm, generate a personnel movement heat map in real time;

[0095] S5. Construct a secure safe to encrypt and store data in slices to ensure data security, deploy federated learning middleware to support feature comparison in multi-party secure computing scenarios and protect data privacy;

[0096] S6. Construct a RESTful API service bus to support on-demand combination of product forms such as video streams, positioning data, and analysis reports, and achieve real-time sharing and exchange of data through API interfaces and data exchange protocols;

[0097] S7. Deeply analyze and mine the stored data, extract valuable information, apply algorithms such as machine learning and deep learning for data analysis, and generate visual reports and warning information;

[0098] S8. Encrypt the transmitted and stored data and implement strict access control;

[0099] S9. Monitor the system operation status in real time, detect and handle anomalies in a timely manner.

[0100] Embodiment 3

[0101] Based on Embodiment 1, a data security sharing system for personnel video images and positioning, the high-altitude acquisition device includes a control device, a housing 101, a support plate 102, multiple groups of propeller power components 103, multiple groups of three-axis pan-tilt heads 104, multiple groups of cameras 105, and an elastic balloon 106. The support plate 102 is installed on the outer wall of the housing 101, multiple groups of propeller power components 103 are all installed on the outer wall of the support plate 102, multiple groups of three-axis pan-tilt heads 104 are all installed at the bottom of the outer wall of the support plate 102, multiple groups of cameras 105 are respectively installed on the mobile ends of multiple groups of three-axis pan-tilt heads 104, the elastic balloon 106 is installed inside the housing 101, and an air vent is provided at the lower part of the outer wall of the housing 101. The control device is connected to the bottom end of the housing 101. The control device is used to control the lifting and moving of the housing 101, and the control device protects multiple groups of cameras 105 by cooperating with the support plate 102;

[0102] The control device includes a power device, a chassis 201, a base 202, a housing 203, a delivery cylinder 204, an air pipe 205, a cable 206, and a sealing ring 207. The base 202 is provided at the bottom end of the chassis 201, the housing 203 is installed at the top end of the chassis 201, the top end of the delivery cylinder 204 is communicated with the housing 203, and the bottom end of the delivery cylinder 204 extends outside the chassis 201. The air pipe 205 and the cable 206 are both connected to the housing 101 at the top end, and the air pipe 205 is communicated with the inside of the elastic balloon 106. The sealing ring 207 is installed on the outer wall of the housing 203. A power device is provided inside the chassis 201. The bottom ends of the air pipe 205 and the cable 206 are connected to the power device, and the air pipe 205 and the cable 206 pass through the inside of the delivery cylinder 204. The power device is used to control the retraction and extension of the air pipe 205 and the cable 206, and the power device is used to provide electricity and hydrogen;

[0103] The power device includes a supply device, a driving device, a cable shaft 301, an electromagnetic slip ring 302, a conveying cylinder 303, a conveying box 304, and two sets of guide wheels 305. The cable shaft 301 is rotatably installed on the inner side wall of the support plate 102. The electromagnetic slip ring 302 is arranged between the support plate 102 and the cable shaft 301. The conveying cylinder 303 is rotatably installed on the inner side wall of the support plate 102. The conveying box 304 is installed on the inner side wall of the support plate 102. The conveying cylinder 303 is rotatably connected to the conveying box 304, and the inside of the conveying cylinder 303 communicates with the inside of the conveying box 304. The conveying box 304 is communicated with the supply device. The supply device is used to convey hydrogen into the conveying box 304. A driving device is arranged inside the support plate 102. The driving device is used to drive the cable shaft 301 and the conveying cylinder 303 to rotate. The two sets of guide wheels 305 are both installed on the inner side wall of the conveying cylinder 204. The bottom end of the cable 206 is connected to the cable shaft 301. The bottom end of the air pipe 205 is communicated with the conveying cylinder 303;

[0104] The driving device includes a support seat 401, a worm 402, a motor 403, and two sets of worm wheels 404. The support seat 401 is installed on the inner side wall of the cover body 101. The worm 402 is rotatably installed on the support seat 401. The motor 403 is installed on the outer side wall of the support seat 401. The output end of the motor 403 is connected to the worm 402. The two sets of worm wheels 404 are respectively installed on the cable shaft 301 and the conveying cylinder 303. The two sets of worm wheels 404 are engaged with the worm 402;

[0105] The supply device includes a storage tank 501, a pump body 502, and a valve 503. The storage tank 501 and the pump body 502 are respectively installed on the inner side wall of the support plate 102. The pump body 502 is connected and arranged between the storage tank 501 and the delivery box 304, and the valve 503 is connected and arranged on the delivery end of the pump body 502; hydrogen is filled into the elastic balloon 106 through the control device. After the elastic balloon 106 expands, the air in the cover body 101 is squeezed out, so that the cover body 101 generates buoyancy and moves upward. The cover body 101 drives multiple groups of cameras 105 to rise to a high place, so that multiple groups of cameras 105 take aerial pictures of the images of people, reducing the occlusion of obstacles and at the same time reducing the number of cameras 105 used. By using hydrogen as the rising power, the energy consumption of the equipment is reduced, and the aerial shooting time is increased. The position of the cover body 101 is corrected by multiple groups of propeller power components 103. By arranging multiple groups of three-axis gimbals 104, the shooting stability of the cameras 105 is improved. By extracting the hydrogen in the elastic balloon 106, it is convenient to store the cover body 101 downward. After the cover body 101 moves downward, multiple groups of cameras 105 are protected by the cooperation of the support plate 102 and the control device, so as to reduce the damage of multiple groups of cameras 105; the power device pays out the air pipe 205 and the cable 206, and at the same time the power device supplies power to the cable 206, conveys hydrogen into the air pipe 205, and the air pipe 205 conveys hydrogen into the elastic balloon 106, so that the cover body 101 rises. The cable 206 supplies power to multiple groups of propeller power components 103, multiple groups of three-axis gimbals 104, and multiple groups of cameras 105, so as to improve the convenience of image shooting power supply. When it rains outdoors, rainwater enters the delivery cylinder 204 through the housing 203, and the rainwater is downwardly diverted and discharged through the delivery cylinder 204, so as to avoid the accumulation of water in the housing 203. After the cover body 101 descends, the support plate 102 and the housing 203 are closed. At this time, multiple groups of three-axis gimbals 104 and multiple groups of cameras 105 are stored inside the housing 203, so as to improve its protection effect; the support plate 102 supplies power to the cable shaft 301 through the electromagnetic slip ring 302, so that the cable shaft 301 supplies power to the cable 206. The supply device supplies hydrogen into the delivery cylinder 303 through the delivery box 304, and the delivery cylinder 303 supplies hydrogen to the air pipe 205. The cable shaft 301 and the delivery cylinder 303 are driven to rotate by the driving device, so that the cable shaft 301 and the delivery cylinder 303 simultaneously wind or pay out the air pipe 205 and the cable 206, so that the cover body 101 drives multiple groups of cameras 105 to move up and down for adjustment. The air pipe 205 and the cable 206 are guided and supported by two groups of guide wheels 305, so as to avoid the abrasion between the air pipe 205 and the cable 206 and the delivery cylinder 204 and improve the protection effect.

[0106] Such as Figures 1 to 11As shown in the figure, a data security sharing method and system for personnel video images and positioning according to the present invention, when working, fills hydrogen into the elastic balloon 106 through a control device. After the elastic balloon 106 expands, it squeezes out the air in the cover body 101, thereby causing the cover body 101 to generate buoyancy and move upward. The cover body 101 drives a plurality of cameras 105 to rise to a high place, so that the plurality of cameras 105 take aerial pictures of the image information of personnel. The position of the cover body 101 is corrected by a plurality of propeller power assemblies 103. By extracting the hydrogen in the elastic balloon 106, it is convenient to store the cover body 101 downward. After the cover body 101 moves downward, a plurality of cameras 105 are protected through the cooperation of the support plate 102 and the control device.

[0107] The main functions achieved by the present invention are as follows:

[0108] 1. By mounting a high-definition camera on a mobile device and a GPS positioning device carried by personnel, video images and positioning data are collected in real time to ensure the timeliness and comprehensiveness of the data and improve the shooting stability of the picture;

[0109] 2. The SLAM technology is adopted to realize three-dimensional scene modeling, and the UWB / Bluetooth beacon is combined to realize passive positioning, generating high-precision three-dimensional scene and personnel position information;

[0110] 3. Based on the multi-target tracking algorithm of particle filter, a personnel movement heat map is generated in real time to provide dynamic monitoring ability for safety management;

[0111] 4. The system has high flexibility and adaptability and can provide customized services according to different scenarios and requirements.

[0112] The propeller power assembly 103, three-axis gimbal 104, camera 105, electromagnetic slip ring 302, motor 403 and pump body 502 of a data security sharing method and system for personnel video images and positioning according to the present invention are purchased on the market. Those skilled in the industry only need to install and operate according to the attached user manual, and there is no need for those skilled in the art to make creative efforts.

[0113] The above are only the preferred embodiments of the present invention. It should be noted that for those of ordinary skill in the art in the technical field, without departing from the technical principle of the present invention, several improvements and modifications can be made, and these improvements and modifications should also be regarded as the protection scope of the present invention.

Claims

1. A data security sharing system for personnel video images and positioning, characterized in that: It includes video image acquisition unit, data preprocessing unit, intelligent analysis and feature extraction unit, dynamic 3D reconstruction and positioning unit, secure storage and privacy computing unit, shared service scheduling unit and data analysis and mining unit; The video image acquisition unit is deployed in the monitoring area. It uses a high-definition camera mounted on a high-altitude acquisition device and a GPS positioning device carried by personnel to collect personnel video images and positioning data in real time, and then transmits the collected raw data to the pre-processing unit. The data preprocessing unit performs cleaning, denoising, and format conversion preprocessing operations on the collected original video images and positioning data, and ensures data quality through a combination of automated algorithms and manual review; The intelligent analysis and feature extraction unit receives data from the data preprocessing unit, integrates a multimodal feature extraction engine, builds an abnormal behavior detection model based on deep learning, outputs spatiotemporal feature vectors and behavior labels, and is used to provide behavior trajectory features to the three-dimensional reconstruction unit and submit feature indexes to the secure storage unit; The dynamic 3D reconstruction and positioning unit receives the feature data of the intelligent analysis unit, uses SLAM technology to achieve 3D scene modeling, combines UWB / Bluetooth beacons to achieve sensorless positioning, and uses a multi-target tracking algorithm based on particle filtering to generate a heat map of personnel movement in real time, providing a spatiotemporal trajectory data package for the shared service unit; The secure storage and privacy computing unit is used to provide basic storage services for all units. By building a secure safe, it implements data sharding and encrypted storage, deploys federated learning middleware, and supports feature comparison in multi-party secure computing scenarios. The shared service scheduling unit is used to build a RESTful API service bus, which supports on-demand combination of video streams, location data, and analytical report product forms. By developing a QoS quality assurance engine, it can achieve adaptive bandwidth adjustment from 64Kbps to 8Mbps, which helps to ensure that users can get a stable and smooth service experience in different network environments. Through the API interface and data exchange protocol, real-time data sharing and exchange can be achieved; The data analysis and mining unit conducts in-depth analysis and mining of stored data, extracts valuable information, applies machine learning, deep learning and other algorithms to perform data analysis, and generates visual reports and early warning information.

2. A data security sharing system for personnel video images and positioning as claimed in claim 1, characterized in that: The shared service scheduling unit also includes a service orchestration engine unit for combining and orchestrating multiple services in the shared service scheduling unit according to specific logic and processes. Through the service orchestration engine, the unit can combine and encapsulate different service units according to business needs to form a complete business process and service scenario.

3. A data security sharing system for personnel video images and positioning as claimed in claim 1, characterized in that: The data preprocessing unit is composed of the following steps by combining automated algorithms and manual review:

1. Automated Algorithms: 1.

1. Use interpolation, mean filling and deletion of missing values ​​to handle missing items in the data, apply statistical methods and machine learning algorithms to identify and handle outliers, and remove duplicate data through database indexing methods to clean the data; 1.

2. Apply filtering technology to remove noise in the image and use smoothing algorithms to reduce jitter and errors in positioning data; 1.

3. Use FFmpeg to convert videos to a unified MP4 format and images to a unified JPEG format; 1.

4. Set quality indicators for image clarity and positioning accuracy, and use automated algorithms for preliminary evaluation. According to the quality indicators, classify the data into three categories: high quality, medium quality, and low quality, and mark them accordingly; 2. Manual review: 2.

1. Data marked as low quality or questionable by the automated algorithm will be assigned to reviewers for manual review, including image clarity and positioning accuracy; 2.

2. Use specialized audit tools or interfaces to facilitate auditors to view, process and mark data, and allow auditors to provide feedback on the results of automated algorithms in order to continuously optimize algorithm performance; 2.

3. Provide professional training to auditors to ensure they are familiar with audit standards and processes, and regularly review audit results to ensure consistency in audit quality; 2.

4. Based on the feedback from auditors and audit results, continuously optimize the automation algorithm to improve the accuracy and efficiency of data preprocessing.

4. A data security sharing system for personnel video images and positioning as claimed in claim 1, characterized in that: The video image acquisition unit also includes a jitter compensation system for performing jitter compensation on the high-definition camera carried by the high-altitude acquisition equipment, thereby enhancing the effect of the high-definition camera being able to stably acquire video images when shooting in the monitoring area; The jitter compensation system includes a jitter detection unit, a jitter compensation unit, an image stabilization unit and a data transmission unit; The jitter detection unit uses the jitter detection algorithm to analyze the IMU data, identify the direction and amplitude of the jitter, and combine it with the GPS data to determine whether the jitter is caused by external factors; The jitter compensation unit uses a compensation control algorithm to generate a three-axis gimbal adjustment command based on the jitter detection results, and adjusts the camera angle in real time to offset the jitter; The image stabilization unit post-processes the video frames and uses an image stabilization algorithm to eliminate residual jitter; The data transmission unit transmits the stabilized video image and positioning data to the pre-processing unit.

5. A data security sharing method for personnel video images and positioning, characterized in that: The following steps are involved: S1. Deploy high-altitude data collection equipment in the monitoring area, equipped with high-definition cameras to collect real-time video images of personnel. Personnel carry GPS positioning equipment with them to collect positioning data in real time; S2, cleaning the original video image, removing blur, jitter and other low-quality frames, denoising the positioning data, and removing outliers; S3, integrated multimodal feature extraction engine, extracts spatiotemporal features from video images and positioning data, builds an abnormal behavior detection model based on deep learning, outputs spatiotemporal feature vectors and behavior labels, sends behavior trajectory features to the three-dimensional reconstruction unit, and submits feature indexes to the secure storage unit; S4, using SLAM technology to achieve scene 3D modeling, generate high-precision 3D scenes, combined with UWB / Bluetooth beacons, achieve non-sensing positioning, accurately track personnel positions, and use particle filter-based multi-target tracking algorithms to generate personnel movement heat maps in real time; S5. Build a safe, encrypt and store data in shards to ensure data security, deploy federated learning middleware, support feature comparison in multi-party secure computing scenarios, and protect data privacy; S6. Build a RESTful API service bus to support on-demand combination of video streams, location data, analysis reports and other product forms, and realize real-time sharing and exchange of data through API interfaces and data exchange protocols; S7. Conduct in-depth analysis and mining of stored data, extract valuable information, apply machine learning, deep learning and other algorithms to perform data analysis, and generate visual reports and early warning information; S8. Encrypt transmitted and stored data and implement strict access control; S9. Monitor the system operation status in real time to detect and handle abnormalities in a timely manner.

6. A data security sharing system for personnel video images and positioning as claimed in claim 1, characterized in that: The high-altitude data acquisition device comprises a control device, a cover (101), a support plate (102), multiple propeller power assemblies (103), multiple three-axis pan-tilt platforms (104), multiple cameras (105) and an elastic balloon (106). The support plate (102) is mounted on the outer wall of the cover (101), the multiple propeller power assemblies (103) are all mounted on the outer wall of the support plate (102), the multiple three-axis pan-tilt platforms (104) are all mounted on the bottom of the outer wall of the support plate (102), the multiple cameras (105) are respectively mounted on the mobile ends of the multiple three-axis pan-tilt platforms (104), the elastic balloon (106) is mounted inside the cover (101), and a vent is provided at the lower part of the outer wall of the cover (101). The control device is connected to the bottom end of the cover (101), the control device is used to control the lifting and lowering movement of the cover (101), and the control device protects the multiple cameras (105) by cooperating with the support plate (102).

7. A data security sharing system for personnel video images and positioning as claimed in claim 6, characterized in that: The control device comprises a power device, a chassis (201), a base (202), a shell (203), a conveying cylinder (204), an air pipe (205), a cable (206) and a sealing ring (207). The chassis (201) is provided with a base (202) at the bottom end, the shell (203) is installed at the top end of the chassis (201), the top end of the conveying cylinder (204) is communicated with the shell (203), the bottom end of the conveying cylinder (204) extends out of the chassis (201), and the tops of the air pipe (205) and the cable (206) are connected to each other. The ends are connected to the cover body (101), and the air pipe (205) is connected to the elastic balloon (106). The sealing ring (207) is installed on the outer wall of the shell (203). A power device is arranged in the chassis (201). The bottom ends of the air pipe (205) and the cable (206) are connected to the power device, and the air pipe (205) and the cable (206) pass through the inside of the conveying cylinder (204). The power device is used to control the retraction and extension of the air pipe (205) and the cable (206), and the power device is used to provide electricity and hydrogen.

8. A data security sharing system for personnel video images and positioning as claimed in claim 7, characterized in that: The power device comprises a supply device, a driving device, a cable shaft (301), an electromagnetic slip ring (302), a conveying cylinder (303), a conveying box (304) and two sets of guide wheels (305); the cable shaft (301) is rotatably mounted on the inner wall of the support plate (102); the electromagnetic slip ring (302) is arranged between the support plate (102) and the cable shaft (301); the conveying cylinder (303) is rotatably mounted on the inner wall of the support plate (102); the conveying box (304) is installed on the inner wall of the support plate (102); the conveying cylinder (303) and the conveying box (304) are rotatably mounted on the inner wall of the support plate (102); The box (304) is rotatably connected, and the conveying cylinder (303) is communicated with the conveying box (304), the conveying box (304) is communicated with the supply device, and the supply device is used to transport hydrogen to the conveying box (304). A driving device is arranged in the support plate (102), and the driving device is used to drive the cable shaft (301) and the conveying cylinder (303) to rotate. Two sets of guide wheels (305) are installed on the inner wall of the conveying cylinder (204), the bottom end of the cable (206) is connected to the cable shaft (301), and the bottom end of the air pipe (205) is communicated with the conveying cylinder (303).

9. A data security sharing system for personnel video images and positioning as claimed in claim 8, characterized in that: The driving device comprises a support seat (401), a worm (402), a motor (403) and two groups of worm gears (404); the support seat (401) is mounted on the inner wall of the cover body (101); the worm (402) is rotatably mounted on the support seat (401); the motor (403) is mounted on the outer wall of the support seat (401); the output end of the motor (403) is connected to the worm (402); the two groups of worm gears (404) are respectively mounted on the cable shaft (301) and the conveying cylinder (303); and the two groups of worm gears (404) are meshed with the worm (402).

10. A data security sharing system for personnel video images and positioning as claimed in claim 8, characterized in that: The supply device comprises a storage tank (501), a pump body (502) and a valve (503); the storage tank (501) and the pump body (502) are respectively mounted on the inner wall of the support plate (102); the pump body (502) is arranged in communication between the storage tank (501) and the delivery box (304); and the valve (503) is arranged in communication on the delivery end of the pump body (502).

Images