On-site law enforcement-based portable recorder data management method, system and platform

Abstract

The invention belongs to the technical field of law enforcement recorders, and discloses a portable recorder data management method, system and platform based on field law enforcement. The method comprises the following steps: collecting and processing multi-dimensional law enforcement data of a portable recorder to generate a law enforcement data sequence; and then encoding the data sequence to generate a plurality of feature tensors, and performing end-to-end integrity verification optimization to obtain an optimized law enforcement low-dimensional tensor. And mapping the optimized law enforcement low-dimensional tensor to a block chain distributed storage space to form the distributed storage space. When a user needs to retrieve data, a user retrieval behavior record is obtained, dynamic behavior characteristics are mined, a retrieval behavior portrait is constructed, unauthorized access behavior analysis and self-adaptive permission diagnosis are carried out, and a data retrieval permission strategy is generated. And carrying out dynamic access control on the multi-level encryption distributed storage space based on the strategy to realize tamper-proof management of law enforcement data. And the security and reliability of law enforcement data are comprehensively ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of law enforcement recorders, and more specifically, to a method, system and platform for managing data of a portable recorder based on on-site law enforcement. Background Art

[0002] With the increasing demand for standardization and informatization of modern law enforcement, portable recorders, as an important tool for on-site law enforcement, are widely used in law enforcement fields such as public security, transportation, and urban management to record the law enforcement process, collect evidence, and ensure the transparency and legality of law enforcement. Portable recorders can collect multi-dimensional law enforcement data in real time, including video, audio, location information, etc., providing an important basis for law enforcement supervision, judicial evidence collection, and post-event analysis.

[0003] However, the existing technology has many problems in the management of body-worn recorder data in on-site law enforcement, which seriously restricts the reliability and application efficiency of law enforcement data. First, the data quality problem is prominent. For example, in actual law enforcement scenarios, the video collected by the body-worn recorder in complex environments (such as insufficient light, noise interference or unstable signal) may be blurred, the audio may be mixed with a lot of background noise, and the location data may be deviated or even lost due to weak GPS signals. These problems directly lead to the insufficient availability of law enforcement data in judicial evidence collection or post-analysis. Secondly, data storage and security issues are particularly severe. Traditional centralized storage methods are easy to become targets of attack, and there is a risk of data tampering, forgery or illegal access. For example, some criminals may modify key video clips through technical means to cover up illegal acts, and law enforcement departments are difficult to prove the originality of data in the absence of an effective anti-tampering mechanism. In addition, data integrity and credibility are not guaranteed enough. In actual operations, law enforcement data may be lost or interrupted during transmission or storage due to network interruption, equipment failure or human error. The lack of effective integrity verification and credible proof mechanism makes it difficult for data to be used as effective evidence in court.

[0004] In addition, the lack of a user access rights management mechanism is also a major pain point. Existing systems often use a static permission allocation method and lack the ability to dynamically monitor user access behavior and detect anomalies. For example, some internal personnel may exploit permission loopholes to frequently access irrelevant data for illegal purposes, and the system cannot identify and prevent such behavior in a timely manner. Finally, low data access efficiency coexists with insufficient privacy protection. In actual law enforcement data call scenarios, authorized users may not be able to quickly obtain key data due to slow system response or unreasonable permission allocation, and sensitive law enforcement data (such as video clips involving personal privacy) may be improperly leaked during the retrieval process due to lack of desensitization, causing legal and moral disputes. The existence of these problems not only reduces the efficiency and security of law enforcement data management, but also weakens the credibility and judicial effectiveness of law enforcement work to a certain extent.

[0005] In view of this, the present invention proposes a body-worn recorder data management method, system and platform based on on-site law enforcement to solve the above problems. Summary of the invention

[0006] In order to overcome the above-mentioned defects of the prior art and to achieve the above-mentioned purpose, the present invention provides the following technical solution: a method for managing data of a portable recorder based on on-site law enforcement, comprising: Step S1: Collect multi-dimensional law enforcement data from the portable recorder; perform quality assessment and filtering on the multi-dimensional law enforcement data from the portable recorder, and perform spatiotemporal label serialization processing to obtain a multi-dimensional law enforcement data sequence; Step S2: Adaptively compress and encode the multi-dimensional law enforcement data sequence to generate multiple low-dimensional law enforcement feature tensors; perform end-to-end integrity verification optimization on the multiple low-dimensional law enforcement feature tensors to generate optimized law enforcement low-dimensional tensors; Step S3: Perform blockchain distributed storage mapping on the optimized law enforcement low-dimensional tensor to obtain a distributed storage space; perform multi-level encryption and obfuscation on the distributed storage space to obtain a multi-level encrypted distributed storage space; Step S4: obtaining the user's behavior record of retrieving law enforcement data; mining the dynamic behavior characteristics of the user's behavior record of retrieving law enforcement data, and constructing a user's behavior profile; Step S5: Analyze the unauthorized access behavior of the user's access behavior profile, make adaptive permission diagnosis decisions, and build a data access permission strategy; Step S6: Perform dynamic access control on the multi-level encrypted distributed storage space based on the data retrieval permission strategy, and build a tamper-proof management model for law enforcement data.

[0007] Preferably, step S1 specifically comprises the following steps: Step S11: Collecting multi-dimensional law enforcement data from the portable recorder; Step S12: performing data type identification and classification processing on the multi-dimensional law enforcement data of the portable recorder to obtain video data, audio data and location data; Step S13: Perform integrity check on the video data, audio data and location data to identify abnormal interruption points; Step S14: performing intelligent compensation processing on the video data, audio data and location data according to the abnormal interruption point to obtain compensated law enforcement data; Step S15: performing quality assessment and filtering on the compensation law enforcement data to generate optimized law enforcement data; Step S16: performing spatiotemporal label serialization processing on the optimized law enforcement data to obtain a multi-dimensional law enforcement data sequence; The specific steps of step S15 are: Perform video clarity evaluation on the compensated law enforcement data to obtain video quality parameters; Performing audio noise analysis on the compensated law enforcement data to obtain audio quality parameters; Conducting location accuracy assessment on the compensated law enforcement data to obtain location quality parameters; A comprehensive quality score is calculated based on the video quality parameter, the audio quality parameter and the location quality parameter to generate a data quality score range; Identify low-quality data of the compensatory law enforcement data based on the data quality score range and mark low-quality data segments; Low-quality data segments are enhanced and optimized to generate optimized law enforcement data.

[0008] Preferably, step S2 specifically comprises the following steps: Step S21: performing multimodal decomposition on the multidimensional law enforcement data sequence to generate a video sequence, an audio sequence and a position sequence; Step S22: performing feature analysis on the video sequence, the audio sequence and the position sequence to obtain a plurality of law enforcement data features; Step S23: Adaptively compress and encode based on multiple law enforcement data features to generate multiple low-dimensional law enforcement feature tensors; Step S24: performing hash integrity calculation on multiple low-dimensional law enforcement feature tensors to obtain law enforcement tensor hash values; Step S25: performing spatiotemporal synchronous standardization on multiple low-dimensional law enforcement feature tensors to generate multiple standardized low-dimensional law enforcement feature tensors; Step S26: Based on the law enforcement tensor hash value, multiple standardized low-dimensional law enforcement feature tensors are end-to-end verified and optimized to generate an optimized law enforcement low-dimensional tensor.

[0009] Preferably, step S3 specifically comprises the following steps: Step S31: performing asymmetric encryption processing on the optimized law enforcement low-dimensional tensor to obtain an encrypted law enforcement tensor; Step S32: Perform blockchain distributed storage mapping on the encrypted law enforcement tensor to obtain a distributed storage space; Step S33: Performing space-time proof generation processing on the distributed storage space to obtain a credible proof chain for law enforcement data; Step S34: Perform multi-level encryption and obfuscation on the trusted proof chain of law enforcement data to obtain a multi-level encrypted distributed storage space.

[0010] Preferably, the specific steps of step S4 are: Step S41: monitoring the entire process of the user retrieving law enforcement data, and obtaining the user's behavior record of retrieving law enforcement data; Step S42: performing time series analysis on the user's behavior record of retrieving law enforcement data to generate time series data of the retrieval behavior; Step S43: mining dynamic behavior features on the retrieval behavior time series data to generate dynamic retrieval behavior features; Step S44: Perform global portrait fitting on the dynamic retrieval behavior features to construct a user retrieval behavior portrait.

[0011] Preferably, the specific steps of step S43 are: Extracting multiple call behavior timestamps based on the call behavior time series data; Calculate the frequency of the user's law enforcement data retrieval behavior records according to multiple retrieval behavior timestamps to obtain the retrieval behavior frequency; Perform data viewing time statistics on retrieved behavior time series data to obtain the viewing time of each section of law enforcement data; According to the frequency of retrieval behavior, the retrieval interval regularity analysis is performed on the viewing time of each section of law enforcement data to obtain the retrieval behavior regularity characteristic data; Identify the type of law enforcement data retrieved by users based on the retrieval behavior time series data; Performing a resource retrieval distribution analysis based on the type of law enforcement data retrieved by the user to obtain retrieval resource distribution data; Dynamic behavior feature analysis is performed on the resource distribution data and the regular feature data of the calling behavior to generate dynamic calling behavior features.

[0012] Preferably, the specific steps of step S5 are: Step S51: Detect abnormal access behavior of the user access behavior profile and extract potential abnormal access behavior data; Step S52: Tracing the entire process of the user's retrieval of law enforcement data to obtain the entire retrieval process; Step S53: performing unauthorized access behavior analysis on potential abnormal access behavior data based on the full link of access behavior to obtain unauthorized access behavior data; Step S54: Performing a security risk quantitative assessment on the unauthorized access behavior data to generate an access behavior risk assessment value; Step S55: Make an adaptive permission diagnosis decision based on the retrieval behavior risk assessment value and build a data retrieval permission strategy.

[0013] Preferably, the specific steps of step S6 are: Step S61: constructing a data tamper-proof architecture for a multi-level encrypted distributed storage space to generate a law enforcement data tamper-proof framework; Step S62: Mapping the authority decision rules of the law enforcement data tamper-proof framework based on the data access authority policy to construct the law enforcement data tamper-proof framework; Step S63: Perform dynamic access control on the law enforcement data authority tamper-proof framework and build a law enforcement data tamper-proof management model; The dynamic access control specifically includes: identifying the access permission level of the current user based on the data access permission policy; The access permission levels include: fully authorized user, partially authorized user and unauthorized user; When the user's access permission level is unauthorized user, the law enforcement data permission anti-tampering framework denies the user from accessing law enforcement data; When the user's access permission level is a partially authorized user, the law enforcement data permission anti-tampering framework allows access after desensitizing the law enforcement data; When the user's access permission level is fully authorized, the law enforcement data permission tamper-proof framework allows the user to fully access law enforcement data and record the chain of evidence of the access behavior.

[0014] A portable recorder data management system based on on-site law enforcement, which is used to implement the portable recorder data management method based on on-site law enforcement, comprises: a data acquisition module, used to collect multi-dimensional law enforcement data from the portable recorder; quality assessment and filtering of the multi-dimensional law enforcement data from the portable recorder, and spatiotemporal label serialization processing to obtain a multi-dimensional law enforcement data sequence; An optimization module is used to adaptively compress and encode multi-dimensional law enforcement data sequences to generate multiple low-dimensional law enforcement feature tensors; perform end-to-end integrity verification optimization on multiple low-dimensional law enforcement feature tensors to generate optimized law enforcement low-dimensional tensors; The encrypted storage module is used to perform distributed storage mapping of the optimized law enforcement low-dimensional tensor on the blockchain to obtain a distributed storage space; perform multi-level encryption and obfuscation on the distributed storage space to obtain a multi-level encrypted distributed storage space; The portrait fitting module is used to obtain the user's behavior record of retrieving law enforcement data; perform dynamic behavior feature mining on the user's behavior record of retrieving law enforcement data, and construct a user's retrieval behavior portrait; The permission construction module is used to analyze the unauthorized access behavior of the user's access behavior profile, make adaptive permission diagnosis decisions, and build data access permission policies; The comprehensive management module performs dynamic access control on multi-level encrypted distributed storage space based on data retrieval permission strategies, and builds a tamper-proof management model for law enforcement data.

[0015] A body-worn recorder data management platform based on on-site law enforcement, the body-worn recorder data management platform based on on-site law enforcement includes a memory and a processor, the memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, the processor executes the body-worn recorder data management method based on on-site law enforcement.

[0016] The technical effects and advantages of the portable recorder data management method, system and platform based on on-site law enforcement of the present invention are as follows: The present invention collects multi-dimensional law enforcement data from body recorders and performs quality assessment filtering and spatiotemporal label serialization processing, thereby solving the problems of unstable quality and weak spatiotemporal correlation of law enforcement data; solves the problems of low storage efficiency and difficulty in ensuring integrity of law enforcement data through adaptive compression coding and end-to-end integrity verification optimization; solves the problems of easy tampering and insufficient security of law enforcement data through blockchain distributed storage mapping and multi-level encryption obfuscation; solves the problems of extensive access control and imprecise authority management of law enforcement data through user retrieval behavior analysis and authority diagnosis decision-making; and realizes the security management of law enforcement data throughout its life cycle by constructing a tamper-proof management model for law enforcement data, effectively improving the credibility and evidentiary effectiveness of law enforcement data. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 A schematic diagram of a portable recorder data management method based on on-site law enforcement according to the present invention; Figure 2 A schematic flow chart of the implementation steps for obtaining a multi-dimensional law enforcement data sequence according to the present invention; Figure 3 A flowchart of the implementation steps for generating an optimized law enforcement low-dimensional tensor for the present invention; Figure 4 A schematic flow chart of the implementation steps of obtaining a multi-level encrypted distributed storage space according to the present invention; Figure 5 The present invention is a schematic diagram of a portable recorder data management system based on on-site law enforcement. DETAILED DESCRIPTION

[0018] 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.

[0019] See also Figures 1 to 4 As shown, this embodiment provides a method for managing body-worn recorder data based on on-site law enforcement, comprising the following steps: Step S1: Collect multi-dimensional law enforcement data from the portable recorder; perform quality assessment and filtering on the multi-dimensional law enforcement data from the portable recorder, and perform spatiotemporal label serialization processing to obtain a multi-dimensional law enforcement data sequence; Step S2: Adaptively compress and encode the multi-dimensional law enforcement data sequence to generate multiple low-dimensional law enforcement feature tensors; perform end-to-end integrity verification optimization on the multiple low-dimensional law enforcement feature tensors to generate optimized law enforcement low-dimensional tensors; Step S3: Perform blockchain distributed storage mapping on the optimized law enforcement low-dimensional tensor to obtain a distributed storage space; perform multi-level encryption and obfuscation on the distributed storage space to obtain a multi-level encrypted distributed storage space; Step S4: obtaining the user's behavior record of retrieving law enforcement data; mining the dynamic behavior characteristics of the user's behavior record of retrieving law enforcement data, and constructing a user's behavior profile; Step S5: Analyze the unauthorized access behavior of the user's access behavior profile, make adaptive permission diagnosis decisions, and build a data access permission strategy; Step S6: Perform dynamic access control on the multi-level encrypted distributed storage space based on the data retrieval permission strategy, and build a tamper-proof management model for law enforcement data.

[0020] The present invention cleans and processes law enforcement data through quality assessment filtering and spatiotemporal label serialization processing, improves data quality, and provides a reliable data basis for subsequent analysis. After obtaining a multi-dimensional law enforcement data sequence, it can better understand the characteristics and changing laws of law enforcement data, provide a basis for subsequent processing and analysis, generate multiple low-dimensional law enforcement feature tensors to reduce data dimensions, improve computing efficiency, and retain important information at the same time. It improves model performance and data expression capabilities through end-to-end integrity verification optimization, and provides better data support for subsequent tasks. Blockchain distributed storage mapping maps data to a new space to improve data security and traceability. Multi-level encryption obfuscation increases data complexity, improves data security, and effectively prevents information leakage and tampering. Dynamic behavior Feature mining deeply understands the user retrieval behavior pattern and constructs the user retrieval behavior portrait, which provides a basis for subsequent analysis and permission diagnosis. The establishment of user retrieval behavior portrait helps law enforcement departments better understand user needs and behavioral habits, and provides data support for law enforcement decisions. Adaptive permission diagnosis decisions timely identify and respond to unauthorized access behaviors, protect the security of law enforcement data, and build data retrieval permission strategies. According to the user retrieval behavior portrait and risk diagnosis results, formulate corresponding protection strategies to improve data security. Dynamic access control realizes real-time monitoring and protection of data according to the data retrieval permission strategy, improves data security and protection capabilities, and builds a law enforcement data tamper-proof management model to comprehensively manage and protect law enforcement data, and improve data security and tamper-proof capabilities.

[0021] In the embodiment of the present invention, refer to Figure 1 , is a schematic flow chart of the steps of a portable recorder data management method based on on-site law enforcement of the present invention. In this example, the steps of the portable recorder data management method based on on-site law enforcement include: Step S1: Collect multi-dimensional law enforcement data from the portable recorder; perform quality assessment and filtering on the multi-dimensional law enforcement data from the portable recorder, and perform spatiotemporal label serialization processing to obtain a multi-dimensional law enforcement data sequence; In this embodiment, various types of law enforcement data are collected and integrated from the body-held recorders worn by law enforcement personnel. These data include: video data (such as law enforcement scene video), audio data (such as on-site conversation records), image data (such as on-site photos), location data (such as GPS positioning information), etc. These heterogeneous multi-dimensional law enforcement data are integrated into a comprehensive body-held recorder multi-dimensional law enforcement data set. Quality assessment and filtering are performed on low-quality data points or noise data in the body-held recorder multi-dimensional law enforcement data set. A quality assessment method based on image clarity or a noise detection model based on machine learning is used to clean and optimize the data. The filtered high-quality data is retained as basic data for subsequent processing. The cleaned multi-dimensional law enforcement data is subjected to spatiotemporal label serialization processing according to timestamp and geographic location information. The spatiotemporal label serialization method adopts a label algorithm based on spatiotemporal correlation, such as spatiotemporal sequence annotation, etc. Through spatiotemporal label serialization, these multi-dimensional law enforcement data are divided into a number of spatiotemporal serialized data sequences. These data sequences reflect the distribution and change patterns of the law enforcement data in the time and space dimensions.

[0022] Step S2: Adaptively compress and encode the multi-dimensional law enforcement data sequence to generate multiple low-dimensional law enforcement feature tensors; perform end-to-end integrity verification optimization on the multiple low-dimensional law enforcement feature tensors to generate optimized law enforcement low-dimensional tensors; In this embodiment, an adaptive compression method is used to encode the multi-dimensional law enforcement data sequence obtained above. An autoencoder model based on deep learning is used to define different encoder structures according to the characteristics of each modal law enforcement data. By training the autoencoder model, the original high-dimensional multi-dimensional law enforcement data is compressed and encoded into a low-dimensional law enforcement feature tensor. The law enforcement data of different modalities are adaptively encoded respectively to generate multiple low-dimensional law enforcement feature tensors. The multiple low-dimensional law enforcement feature tensors generated above are input into an end-to-end optimization framework. This framework adopts the structure of a multimodal fusion network to learn the association and interaction between multiple low-dimensional features. During the optimization process, multiple integrity verification functions are defined, such as hash verification, digital signature verification, blockchain verification, etc., to perform end-to-end integrity verification optimization training. Finally, a set of optimized multimodal low-dimensional law enforcement feature tensors, i.e., optimized law enforcement low-dimensional tensors, are obtained.

[0023] Step S3: Perform blockchain distributed storage mapping on the optimized law enforcement low-dimensional tensor to obtain a distributed storage space; perform multi-level encryption and obfuscation on the distributed storage space to obtain a multi-level encrypted distributed storage space; In this embodiment, the optimized law enforcement low-dimensional tensor is used as input and processed by a blockchain distributed storage mapping module. This blockchain distributed storage mapping module is based on blockchain technology and maps the original low-dimensional law enforcement feature tensor to a distributed storage space. Each data block in the distributed storage space corresponds to a part of the original low-dimensional law enforcement feature tensor, but there is a complex mapping relationship between the two. Through this blockchain distributed mapping, a distributed storage space is obtained, which contains a distributed representation of the original multimodal law enforcement features. The data in the distributed storage space is further subjected to a series of multi-level encryption and obfuscation processing. Through these complex multi-level encryption transformations, the distributed storage space is further obfuscated and protected to generate a multi-level encrypted distributed storage space.

[0024] Step S4: obtaining the user's behavior record of retrieving law enforcement data; mining the dynamic behavior characteristics of the user's behavior record of retrieving law enforcement data, and constructing a user's behavior profile; In this embodiment, the behavior records of users retrieving law enforcement data are collected and integrated from various business systems and applications within the law enforcement department. Machine learning and data mining methods are used to extract the dynamic retrieval behavior characteristics of users based on the collected behavior records of users retrieving law enforcement data. The characteristics include the user's retrieval mode, such as retrieval frequency, retrieval time distribution, retrieval path, etc., the user's operation behavior, such as common functions, preference settings, abnormal operations, etc., and the user's retrieval content, such as case type, regional distribution, time span, etc. These dynamic retrieval behavior characteristics are integrated into a user retrieval behavior portrait model, which can comprehensively describe the retrieval behavior characteristics of each user.

[0025] Step S5: Analyze the unauthorized access behavior of the user's access behavior profile, make adaptive permission diagnosis decisions, and build a data access permission strategy; In this embodiment, the user access behavior portrait model constructed above is input into an unauthorized access behavior analysis module based on machine learning. This analysis module will refer to the known unauthorized access behavior characteristics to compare and analyze the access behavior portrait of each user. The results of the unauthorized access behavior analysis are input into an adaptive permission diagnosis model. This permission diagnosis model will comprehensively consider factors such as user access behavior portraits, unauthorized access characteristics, and law enforcement security policies to dynamically evaluate the user's access permissions. Based on the permission evaluation results, a targeted data access permission management strategy is formulated.

[0026] Step S6: Perform dynamic access control on the multi-level encrypted distributed storage space based on the data retrieval permission strategy, and build a tamper-proof management model for law enforcement data.

[0027] In this embodiment, based on the multi-level encrypted distributed storage space, the overall architecture of law enforcement data tamper prevention is defined. This architecture includes: encrypted storage module, access control module, risk monitoring module, etc. Through the collaborative work of these functional modules, a complete law enforcement data tamper prevention framework is constructed. This framework provides basic support for subsequent law enforcement data tamper prevention. The obtained data retrieval permission management strategy is connected with the law enforcement data tamper prevention framework. According to different permission evaluation values, corresponding access control, risk response and other decision rules are formulated. These rules are mapped to the functional modules of the law enforcement data tamper prevention framework to construct a law enforcement-level permission tamper prevention framework. This permission tamper prevention framework realizes refined law enforcement data access control and tamper warning. The constructed permission tamper prevention framework is deployed to the data management system of the law enforcement department. For each user's law enforcement data retrieval request, the framework will perform real-time access control based on the dynamic permission evaluation results. The means of access control include: dynamic encryption, permission management, risk warning, etc. Through this dynamic access control mechanism, a comprehensive law enforcement data tamper prevention management model is constructed.

[0028] In this embodiment, refer to Figure 2 , is a flowchart of detailed implementation steps of step S1. In this embodiment, the detailed implementation steps of step S1 include: Step S11: Collecting multi-dimensional law enforcement data from the portable recorder; Step S12: performing data type identification and classification processing on the multi-dimensional law enforcement data of the portable recorder to obtain video data, audio data and location data; Step S13: Perform integrity check on the video data, audio data and location data to identify abnormal interruption points; Step S14: performing intelligent compensation processing on the video data, audio data and location data according to the abnormal interruption point to obtain compensated law enforcement data; Step S15: performing quality assessment and filtering on the compensation law enforcement data to generate optimized law enforcement data; Step S16: Perform spatiotemporal label serialization processing on the optimized law enforcement data to obtain a multi-dimensional law enforcement data sequence.

[0029] In this embodiment, various types of law enforcement-related data, including video data, audio data, location data, etc., are collected from the portable recorders worn by law enforcement personnel. The collected raw data are reasonably classified and archived to establish a multi-dimensional law enforcement data set for the portable recorder. The data set is identified by machine learning methods, and the data is divided into video data, audio data, and location data. For video data, visual features are extracted by computer vision technology; for audio data, sound features are extracted by audio processing technology; for location data, spatial features are extracted by geographic information system. The time series in the video data, audio data, and location data are analyzed to detect whether there are abnormal interruption points. Methods based on time series analysis, such as continuity detection and signal mutation detection, are used to identify abnormal interruption points in the data. For the identified abnormal interruption points, intelligent compensation models (such as video frame interpolation, audio signal reconstruction, location trajectory smoothing, etc.) are used to compensate the interruption data. The results of the compensation processing are filled into the original law enforcement data to repair the abnormal interruption data points and obtain compensated law enforcement data. For the compensatory law enforcement data, a quality assessment algorithm is used for filtering, the video data is evaluated for clarity, the audio data is analyzed for noise levels, and the location data is evaluated for accuracy.

[0030] Specifically, the clarity of the video data is evaluated: Calculate the clarity index of video data ; ;in, For video data The gradient amplitude of the frame image, is the noise estimation parameter, which represents the variance estimation value of the noise in the video and is used to quantify the impact of noise on video quality; is the edge clarity evaluation index, and is the weight coefficient, is the number of frames of video data, The number of evaluation regions refers to the number of regions divided in video quality assessment. This regional division strategy can more accurately evaluate the quality status of different parts of the video.

[0031] ;in, and are the horizontal and vertical gradients, respectively. For the assessment area, For the The number of pixels in the evaluation area, For the The edge contrast in the evaluation area is is the standard reference contrast (preset); If the clarity index If it is greater than the preset threshold, the clarity of the corresponding evaluation area does not meet the standard, that is, the quality does not meet the standard.

[0032] The noise level analysis method for audio data adopts the signal-to-noise ratio weighted evaluation method, that is, a weighted function of the signal-to-noise ratio is calculated. The weights in the weighted function are adaptive weights, which are dynamically adjusted according to the importance of the speech. For example, an audio recorded by a law enforcement recorder is a policeman questioning a witness about a traffic accident on a noisy street. The characteristics of the original audio include the voice of the witness describing the accident (key evidence), the voice of the police questioning (procedural content), background traffic noise (interference content) and occasional conversations with passers-by (irrelevant content); first, through voice recognition and semantic analysis, different parts of the voice content are identified. For example, when the witness describes "the red car ran a red light and hit the bicycle", the system identifies it as high-importance content (importance score = 0.9), when the police asks "Excuse me, where were you standing at the time", it is identified as medium-importance content (importance score = 0.6), and when the passerby says "The weather is really good today", it is identified as low-importance content (importance score = 0.2); the voice of the witness describing the accident is given the highest weight, and the clarity of this part of the content will be prioritized when calculating the signal-to-noise ratio and subsequent noise reduction processing. The police questioning is given a medium weight to maintain basic clarity; the conversations with passers-by are given a low weight and may be more strongly suppressed during the noise reduction process.

[0033] Evaluate the accuracy of the location data: The location data quality assessment uses a multi-source fusion accuracy assessment method to calculate the location accuracy credibility ; ;in, for The position coordinates at the time, is the estimated position after multi-source fusion, is the position variance (usually expressed as a covariance matrix), which is a quantitative representation of the uncertainty of the position estimate and reflects the degree to which the estimated position may deviate from the true position; is the sampling interval, is the time attenuation coefficient; is the total sampling time.

[0034] It should be explained that the estimated position after multi-source fusion refers to the more accurate and reliable position coordinates obtained by integrating multiple positioning signal sources. In the law enforcement recorder system, these signal sources may include: GPS / GNSS receivers, base station triangulation positioning, WiFi positioning, Bluetooth beacons, inertial navigation systems (INS), and step count and direction estimation, which are fused through weighted averaging or Kalman filtering; if the position accuracy is reliable, If the accuracy is less than the preset threshold, the corresponding location data does not meet the standard, that is, the quality does not meet the standard.

[0035] For data segments that do not meet the quality standards, data enhancement, noise filtering and other methods are used to optimize and process them to generate optimized law enforcement data. The optimized law enforcement data is serialized with spatiotemporal labels, and accurate timestamps and geographic location labels are added to each segment of data.

[0036] The law enforcement data is serialized using spatiotemporal serialization algorithms (such as spatiotemporal correlation analysis, spatiotemporal clustering, etc.) to obtain a multidimensional law enforcement data sequence. The serialization results are visualized and analyzed to understand the spatiotemporal distribution characteristics and potential patterns of law enforcement data.

[0037] In this embodiment, the specific steps of step S15 are: Perform video clarity evaluation on the compensated law enforcement data to obtain video quality parameters; Performing audio noise analysis on the compensated law enforcement data to obtain audio quality parameters; Conducting location accuracy assessment on the compensated law enforcement data to obtain location quality parameters; A comprehensive quality score is calculated based on the video quality parameter, the audio quality parameter and the location quality parameter to generate a data quality score range; Identify low-quality data of the compensatory law enforcement data based on the data quality score range and mark low-quality data segments; Low-quality data segments are enhanced and optimized to generate optimized law enforcement data.

[0038] In this embodiment, for the compensation law enforcement data of law enforcement, computer vision technology is used to evaluate the clarity of video data, calculate the resolution, contrast, sharpness, etc. of the video, and generate parameters representing the video quality, namely, video quality parameters. For audio data, audio analysis technology is used to evaluate the noise level, calculate indicators such as signal-to-noise ratio and spectrum distribution, and generate parameters representing the audio quality, namely, audio quality parameters; for position data, the GPS positioning accuracy is evaluated, and the stability and accuracy of the position signal (position accuracy credibility) are analyzed to generate parameters representing the position quality, namely, position quality parameters (GPS positioning accuracy, stability, accuracy); specifically, the evaluation process of GPS positioning accuracy includes: collecting the satellite distribution of the GPS receiver, obtaining the corresponding precision factor, and obtaining the position precision factor from the precision factor , based on the location precision factor, the GPS positioning accuracy assessment value is calculated ;in is the standard deviation of the user equivalent range error (User Equivalent Range Error).

[0039] The stability analysis methods of position signal include: Analyze the sampling interval of position data in a continuous time window and calculate the temporal stability index , that is, stability; ;in, For the sampling points, For the sampling points, is the total number of sampling points.

[0040] The video quality parameters, audio quality parameters and location quality parameters are combined, and the comprehensive quality score of each segment of law enforcement data is calculated through weighted average or machine learning model. The qualified range of quality scores is defined for subsequent data screening. For each segment of compensatory law enforcement data, its comprehensive quality score is compared with the predetermined quality score range, and the data segments with scores below the threshold are marked as low-quality data segments. These low-quality data segments may contain blurry videos, noisy audio or inaccurate location information. Targeted enhancement and optimization processing is adopted for the marked low-quality data segments. For video data, super-resolution reconstruction, image enhancement and other technologies are used to improve clarity. For audio data, noise reduction, audio enhancement and other technologies are used to improve sound quality. For location data, trajectory smoothing, position correction and other technologies are used to improve accuracy.

[0041] In this embodiment, refer to Figure 3 , is a flowchart of detailed implementation steps of step S2. In this embodiment, the detailed implementation steps of step S2 include: Step S21: performing multimodal decomposition on the multidimensional law enforcement data sequence to generate a video sequence, an audio sequence and a position sequence; Step S22: performing feature analysis on the video sequence, the audio sequence and the position sequence to obtain a plurality of law enforcement data features; Step S23: Adaptively compress and encode based on multiple law enforcement data features to generate multiple low-dimensional law enforcement feature tensors; Step S24: performing hash integrity calculation on multiple low-dimensional law enforcement feature tensors to obtain law enforcement tensor hash values; Step S25: performing spatiotemporal synchronous standardization on multiple low-dimensional law enforcement feature tensors to generate multiple standardized low-dimensional law enforcement feature tensors; Step S26: Based on the law enforcement tensor hash value, multiple standardized low-dimensional law enforcement feature tensors are end-to-end verified and optimized to generate an optimized law enforcement low-dimensional tensor.

[0042] In this embodiment, a multidimensional law enforcement data sequence (including video data, audio data and position data) is decomposed into data sequences of different modes, and video sequences, audio sequences and position sequences are extracted. The video sequence is subjected to frame analysis, scene recognition and other processing to extract visual features. The audio sequence is subjected to spectrum analysis, speech recognition and other processing to extract audio features. The position sequence is subjected to trajectory analysis, spatial clustering and other processing to extract position features. After feature analysis, a set of data describing the multimodal features of the law enforcement data is obtained. The multiple law enforcement data features obtained in the previous step are input into a deep learning model based on an autoencoder. The model will adaptively learn the potential low-dimensional representation of the data and generate corresponding low-dimensional law enforcement feature tensors. A hash value is calculated for each low-dimensional law enforcement feature tensor as a verification of data integrity. Certificate identification, these hash values ​​reflect the characteristic fingerprints of the data in the compression process, and these law enforcement tensor hash values ​​are used as a new data set for subsequent integrity verification. Due to the differences in the spatiotemporal characteristics of different modal data, it is necessary to perform spatiotemporal synchronization and standardization on the low-dimensional law enforcement feature tensors. Use time alignment, spatial normalization and other methods to map each modal data to a unified spatiotemporal reference system. After standardization, a set of standardized low-dimensional law enforcement feature tensors with spatiotemporal consistency are obtained. The above-mentioned law enforcement tensor hash values ​​and the standardized low-dimensional law enforcement feature tensors are input into an end-to-end integrity verification optimization model. The model will learn how to verify and maintain the integrity of the data, generate a more optimized low-dimensional law enforcement feature tensor, and greatly reduce the dimension and storage overhead of the data while ensuring data integrity.

[0043] In this embodiment, refer to Figure 4 , is a flowchart of detailed implementation steps of step S3. In this embodiment, the detailed implementation steps of step S3 include: Step S31: performing asymmetric encryption processing on the optimized law enforcement low-dimensional tensor to obtain an encrypted law enforcement tensor; Step S32: Perform blockchain distributed storage mapping on the encrypted law enforcement tensor to obtain a distributed storage space; Step S33: Performing space-time proof generation processing on the distributed storage space to obtain a credible proof chain for law enforcement data; Step S34: Perform multi-level encryption and obfuscation on the trusted proof chain of law enforcement data to obtain a multi-level encrypted distributed storage space.

[0044] In this embodiment, multimodal law enforcement data (such as video, audio, location, etc.) are feature fused to generate a low-dimensional tensor representation, and an asymmetric encryption algorithm (such as RSA, ECC, etc.) is used to encrypt the low-dimensional law enforcement tensor to protect the confidentiality of the data. The encrypted law enforcement tensor is output as the input for the next step, and a blockchain distributed storage mapping model is defined to map the encrypted law enforcement tensor to the blockchain network. Based on blockchain technology, a distributed storage space is constructed, and each data block contains encrypted law enforcement data and corresponding verification information. The constructed distributed storage space is output for subsequent proof chain processing, and a spatiotemporal proof generation mechanism is defined to prove the time and space dimensions of data in the distributed storage space, generate a timestamp proof and a location proof for each data block, and link these proof information into a complete proof chain to ensure the spatiotemporal integrity of the data. The law enforcement data trusted proof chain is subjected to multi-level encryption processing, including data layer encryption, proof layer encryption, and access layer encryption. The proof chain is protected through multi-level encryption obfuscation to enhance the security of the data, and the proof chain that has undergone multi-level encryption obfuscation is output as the final multi-level encrypted distributed storage space.

[0045] In this embodiment, step S4 includes the following steps: Step S41: monitoring the entire process of the user retrieving law enforcement data, and obtaining the user's behavior record of retrieving law enforcement data; Step S42: performing time series analysis on the user's behavior record of retrieving law enforcement data to generate time series data of the retrieval behavior; Step S43: mining dynamic behavior features on the retrieval behavior time series data to generate dynamic retrieval behavior features; Step S44: Perform global portrait fitting on the dynamic retrieval behavior features to construct a user retrieval behavior portrait.

[0046] In this embodiment, a user behavior monitoring system is deployed within the law enforcement department to track and record the entire process of users retrieving law enforcement data in real time. The monitoring content includes: user identity information, retrieval time, retrieval object (video, audio, location data, etc.), retrieval operation (viewing, exporting, etc.), etc. The collected user retrieval law enforcement data behavior records are structured and stored to form a user retrieval law enforcement data behavior record set. The timestamp information in the user retrieval law enforcement data behavior record is analyzed and converted into time series data. The time series data is subjected to statistical analysis, time window aggregation, and other processing to generate retrieval behavior time series data that is more rich in time dimension features. These time series data reflect the time change trend of user retrieval behavior. For the time-series user retrieval behavior data, time series analysis, anomaly detection and other methods are used to mine the dynamic features contained therein. Features such as user retrieval frequency, retrieval time distribution, and retrieval object association are extracted. These dynamic features describe the time change pattern and abnormal pattern of user retrieval behavior. The dynamic retrieval behavior features mined above are integrated into a user portrait model. This model is based on statistical analysis or machine learning, and is used to fit and describe the user's global access behavior characteristics. Through the user portrait model, a comprehensive user access behavior portrait is generated, reflecting the user's access mode in multiple dimensions such as time, space, and objects. This user access behavior portrait provides a valuable basis for subsequent user behavior analysis, permission control, and other work.

[0047] In this embodiment, the specific steps of step S43 are: Extracting multiple call behavior timestamps based on the call behavior time series data; Calculate the frequency of the user's law enforcement data retrieval behavior records according to multiple retrieval behavior timestamps to obtain the retrieval behavior frequency; Perform data viewing time statistics on retrieved behavior time series data to obtain the viewing time of each section of law enforcement data; According to the frequency of retrieval behavior, the retrieval interval regularity analysis is performed on the viewing time of each section of law enforcement data to obtain the retrieval behavior regularity characteristic data; Identify the type of law enforcement data retrieved by users based on the retrieval behavior time series data; Performing a resource retrieval distribution analysis based on the type of law enforcement data retrieved by the user to obtain retrieval resource distribution data; Dynamic behavior feature analysis is performed on the resource distribution data and the regular feature data of the calling behavior to generate dynamic calling behavior features.

[0048] In this embodiment, the collected user retrieval behavior time series data is timestamped to obtain the time point of each retrieval event. These timestamps reflect the user's actual retrieval time series. The extracted retrieval behavior timestamps are used to count the number of retrievals of each user within a certain time window. The user's average retrieval frequency is calculated based on the number of retrievals as the retrieval behavior frequency feature. This feature reflects the activeness of the user's retrieval behavior. For each retrieval event, the length of time the user views the law enforcement data is recorded, and the average viewing time of each section of law enforcement data is counted as the data viewing time feature of the user's retrieval behavior. This feature reflects the user's attention to different law enforcement data. Combined with the retrieval frequency and data viewing time data, the regular characteristics of the user's retrieval behavior are analyzed. For example, High-frequency access but short viewing time indicates browsing behavior; low-frequency access but long viewing time indicates in-depth analysis behavior. These regular features of access behavior are integrated into access behavior regular feature data, and the access behavior time series data is analyzed to identify the specific law enforcement data types accessed by users, such as video data, audio data, location data, etc. These access law enforcement data type information is used as an important feature of the user's access behavior. The identified access law enforcement data types are classified, and the distribution of users' access to different types of resources is statistically calculated. This access resource distribution data reflects the user's work focus and attention direction. The access resource distribution data and access behavior regular feature data obtained above are integrated and analyzed, and a machine learning model is used to mine the dynamic patterns and abnormal features of user access behavior.

[0049] In this embodiment, step S5 includes the following steps: Step S51: Detect abnormal access behavior of the user access behavior profile and extract potential abnormal access behavior data; Step S52: Tracing the entire process of the user's retrieval of law enforcement data to obtain the entire retrieval process; Step S53: performing unauthorized access behavior analysis on potential abnormal access behavior data based on the full link of access behavior to obtain unauthorized access behavior data; Step S54: Performing a security risk quantitative assessment on the unauthorized access behavior data to generate an access behavior risk assessment value; Step S55: Make an adaptive permission diagnosis decision based on the retrieval behavior risk assessment value and build a data retrieval permission strategy.

[0050] In this embodiment, the user retrieval behavior portrait model constructed above is used to perform anomaly detection on the real-time user retrieval behavior records of law enforcement data. Statistical learning, anomaly detection and other methods are used to identify retrieval behaviors that deviate greatly from the user's normal retrieval behavior patterns. These retrieval behavior data marked as abnormal are extracted as potential abnormal retrieval behavior data. The collected user retrieval behavior records of law enforcement data, including normal retrieval and abnormal retrieval, are comprehensively traced for the retrieval path. The traced content includes: user identity, retrieval time, retrieval object, retrieval operation, etc. Through this full-link tracing, the entire behavior trajectory of the user from login to retrieval completion is restored. These retrieval behavior full-link data provide a basis for subsequent unauthorized access behavior analysis. The extracted abnormal retrieval behavior data is combined with the obtained retrieval behavior full link for in-depth Analyze and use methods such as permission model recognition and abnormal event correlation to identify existing unauthorized access behaviors, extract the retrieval behavior data that are determined to be unauthorized access as unauthorized retrieval behavior data, quantify the potential security risk level of the unauthorized retrieval behavior data, consider factors such as the type, scope, and impact of unauthorized behaviors, define a security risk assessment model for scoring, and generate a retrieval behavior risk assessment value based on the assessment results as the basis for subsequent permission management and control. Input the risk assessment value of each unauthorized retrieval behavior into an adaptive permission diagnosis model, which will automatically generate corresponding permission diagnosis decisions based on real-time risk conditions. These decisions include: triggering alarms, restricting retrieval permissions, initiating emergency responses, etc. These diagnostic decisions are summarized to form a complete set of data retrieval permission strategies.

[0051] In this embodiment, the specific steps of step S6 are: Step S61: constructing a data tamper-proof architecture for a multi-level encrypted distributed storage space to generate a law enforcement data tamper-proof framework; Step S62: Mapping the authority decision rules of the law enforcement data tamper-proof framework based on the data access authority policy to construct the law enforcement data tamper-proof framework; Step S63: Perform dynamic access control on the law enforcement data authority tamper-proof framework and build a law enforcement data tamper-proof management model; The dynamic access control specifically includes: identifying the access permission level of the current user based on the data access permission policy; The access permission levels include: fully authorized user, partially authorized user and unauthorized user; When the user's access permission level is unauthorized user, the law enforcement data permission anti-tampering framework denies the user from accessing law enforcement data; When the user's access permission level is a partially authorized user, the law enforcement data permission anti-tampering framework allows access after desensitizing the law enforcement data; When the user's access permission level is fully authorized, the law enforcement data permission tamper-proof framework allows the user to fully access law enforcement data and record the chain of evidence of the access behavior.

[0052] In this embodiment, a complete data tamper-proof architecture is defined based on a multi-level encrypted distributed storage space. The architecture includes multiple functions such as a data storage layer, an anti-tamper verification layer, and an access control layer. The data storage layer is responsible for managing encrypted law enforcement data, the anti-tamper verification layer is responsible for verifying the integrity and authenticity of the data, and the access control layer is responsible for managing user access rights. Through the collaborative work of these functional modules, a law enforcement data tamper-proof framework is constructed, and the aforementioned data retrieval permission strategy is integrated with the law enforcement data tamper-proof framework. According to the retrieval permission levels of different users, corresponding access control rules are formulated. These rules include: data access permission control, operation permission control, audit tracking, etc. These rules are mapped to the access control layer of the law enforcement data tamper-proof framework to form a complete law enforcement data tamper-proof framework. According to the permission tamper-proof framework, a dynamic access control mechanism is implemented for the law enforcement data permission tamper-proof framework. For each user request to retrieve law enforcement data, the user's retrieval permission level will be evaluated in real time. Different access control measures will be taken according to the different permission levels: for unauthorized users, their retrieval requests will be directly rejected, and the access denial log will be recorded. For some authorized users, the law enforcement data will be desensitized, such as blurring faces, hiding sensitive information, etc., and then users will be allowed to retrieve the desensitized law enforcement data. For fully authorized users, they are allowed to retrieve the law enforcement data in full, but detailed retrieval behavior logs will be recorded to form a retrieval behavior evidence chain. Through this dynamic access control mechanism, all-round protection of law enforcement data is achieved, and a complete law enforcement data tamper-proof management model is constructed.

[0053] The present invention collects multi-dimensional law enforcement data from body recorders and performs quality assessment filtering and spatiotemporal label serialization processing, thereby solving the problems of unstable quality and weak spatiotemporal correlation of law enforcement data; solves the problems of low storage efficiency and difficulty in ensuring integrity of law enforcement data through adaptive compression coding and end-to-end integrity verification optimization; solves the problems of easy tampering and insufficient security of law enforcement data through blockchain distributed storage mapping and multi-level encryption obfuscation; solves the problems of extensive access control and imprecise authority management of law enforcement data through user retrieval behavior analysis and authority diagnosis decision-making; and realizes the security management of law enforcement data throughout its life cycle by constructing a tamper-proof management model for law enforcement data, effectively improving the credibility and evidentiary effectiveness of law enforcement data.

[0054] The above describes the portable recorder data management method based on on-site law enforcement in the embodiment of the present application. The following describes the portable recorder data management system based on on-site law enforcement in the embodiment of the present application. Please refer to Figure 5In the embodiment of the present application, an embodiment of a portable recorder data management system based on on-site law enforcement includes: The data collection module is used to collect multi-dimensional law enforcement data from the portable recorder; perform quality assessment and filtering on the multi-dimensional law enforcement data from the portable recorder, and perform spatiotemporal label serialization processing to obtain a multi-dimensional law enforcement data sequence; An optimization module is used to adaptively compress and encode multi-dimensional law enforcement data sequences to generate multiple low-dimensional law enforcement feature tensors; perform end-to-end integrity verification optimization on multiple low-dimensional law enforcement feature tensors to generate optimized law enforcement low-dimensional tensors; The encrypted storage module is used to perform distributed storage mapping of the optimized law enforcement low-dimensional tensor on the blockchain to obtain a distributed storage space; perform multi-level encryption and obfuscation on the distributed storage space to obtain a multi-level encrypted distributed storage space; The portrait fitting module is used to obtain the user's behavior record of retrieving law enforcement data; perform dynamic behavior feature mining on the user's behavior record of retrieving law enforcement data, and construct a user's retrieval behavior portrait; The permission construction module is used to analyze the unauthorized access behavior of the user's access behavior profile, make adaptive permission diagnosis decisions, and build data access permission policies; The comprehensive management module performs dynamic access control on multi-level encrypted distributed storage space based on data retrieval permission strategies, and builds a tamper-proof management model for law enforcement data.

[0055] The present application also provides a body-worn recorder data management platform based on on-site law enforcement, the body-worn recorder data management platform based on on-site law enforcement includes a memory and a processor, the memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, the processor executes the steps of the body-worn recorder data management method based on on-site law enforcement in the above-mentioned embodiments.

[0056] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described systems, systems and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0057] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, it is still possible for a person skilled in the art to modify the technical solutions described in the aforementioned embodiments or to replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

[0058] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or device including the element.

[0059] In the description of the present invention, it should be understood that the terms "first", "second", etc. are only used to distinguish the descriptions and cannot be understood as indicating or implying relative importance.

[0060] In the description of the present invention, unless otherwise specified, "plurality" means two or more.

[0061] In the description of the present invention, "several" means one or more than one, and "a large number" means two or more than two.

[0062] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "examples", "specific examples", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0063] The formulas in this manual are all dimensionless and calculated numerically. The formulas are obtained by collecting a large amount of data and performing software simulation to obtain the most recent real situation. The preset parameters and thresholds in the formulas are set by technicians in this field according to actual conditions.

[0064] Although the embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the claims and their equivalents.

Claims

1. A method for managing data of a portable recorder based on on-site law enforcement, characterized in that: include: Step S1: Collect multi-dimensional law enforcement data from the body recorder; Perform quality assessment and filtering on the multi-dimensional law enforcement data of the body recorder, and perform spatiotemporal label serialization processing to obtain a multi-dimensional law enforcement data sequence; Step S2: Adaptively compress and encode the multi-dimensional law enforcement data sequence to generate multiple low-dimensional law enforcement feature tensors; perform end-to-end integrity verification optimization on the multiple low-dimensional law enforcement feature tensors to generate optimized law enforcement low-dimensional tensors; Step S3: Perform blockchain distributed storage mapping on the optimized law enforcement low-dimensional tensor to obtain a distributed storage space; perform multi-level encryption and obfuscation on the distributed storage space to obtain a multi-level encrypted distributed storage space; Step S4: obtaining the user's behavior record of retrieving law enforcement data; mining the dynamic behavior characteristics of the user's behavior record of retrieving law enforcement data, and constructing a user's behavior profile; Step S5: Analyze the unauthorized access behavior of the user's access behavior profile, make adaptive permission diagnosis decisions, and build a data access permission strategy; Step S6: Perform dynamic access control on the multi-level encrypted distributed storage space based on the data retrieval permission strategy, and build a tamper-proof management model for law enforcement data.

2. The method for managing data of a portable recorder based on on-site law enforcement according to claim 1, characterized in that: The specific steps of step S1 are: Step S11: Collecting multi-dimensional law enforcement data from the portable recorder; Step S12: performing data type identification and classification processing on the multi-dimensional law enforcement data of the portable recorder to obtain video data, audio data and location data; Step S13: Perform integrity check on the video data, audio data and location data to identify abnormal interruption points; Step S14: performing intelligent compensation processing on the video data, audio data and location data according to the abnormal interruption point to obtain compensated law enforcement data; Step S15: performing quality assessment and filtering on the compensation law enforcement data to generate optimized law enforcement data; Step S16: performing spatiotemporal label serialization processing on the optimized law enforcement data to obtain a multi-dimensional law enforcement data sequence; The specific steps of step S15 are: Perform video clarity evaluation on the compensated law enforcement data to obtain video quality parameters; Performing audio noise analysis on the compensated law enforcement data to obtain audio quality parameters; Conducting location accuracy assessment on the compensated law enforcement data to obtain location quality parameters; A comprehensive quality score is calculated based on the video quality parameter, the audio quality parameter and the location quality parameter to generate a data quality score range; Identify low-quality data of the compensatory law enforcement data based on the data quality score range and mark low-quality data segments; Low-quality data segments are enhanced and optimized to generate optimized law enforcement data.

3. The method for managing data of a portable recorder based on on-site law enforcement according to claim 2, characterized in that: The specific steps of step S2 are: Step S21: performing multimodal decomposition on the multidimensional law enforcement data sequence to generate a video sequence, an audio sequence and a position sequence; Step S22: performing feature analysis on the video sequence, the audio sequence and the position sequence to obtain a plurality of law enforcement data features; Step S23: Adaptively compress and encode based on multiple law enforcement data features to generate multiple low-dimensional law enforcement feature tensors; Step S24: performing hash integrity calculation on multiple low-dimensional law enforcement feature tensors to obtain law enforcement tensor hash values; Step S25: performing spatiotemporal synchronous standardization on multiple low-dimensional law enforcement feature tensors to generate multiple standardized low-dimensional law enforcement feature tensors; Step S26: Based on the law enforcement tensor hash value, multiple standardized low-dimensional law enforcement feature tensors are end-to-end verified and optimized to generate an optimized law enforcement low-dimensional tensor.

4. The method for managing data of a portable recorder based on on-site law enforcement according to claim 3, characterized in that: The specific steps of step S3 are: Step S31: performing asymmetric encryption processing on the optimized law enforcement low-dimensional tensor to obtain an encrypted law enforcement tensor; Step S32: Perform blockchain distributed storage mapping on the encrypted law enforcement tensor to obtain a distributed storage space; Step S33: Performing space-time proof generation processing on the distributed storage space to obtain a credible proof chain for law enforcement data; Step S34: Perform multi-level encryption and obfuscation on the trusted proof chain of law enforcement data to obtain a multi-level encrypted distributed storage space.

5. The method for managing data of a portable recorder based on on-site law enforcement according to claim 4, characterized in that: The specific steps of step S4 are: Step S41: monitoring the entire process of the user retrieving law enforcement data, and obtaining the user's behavior record of retrieving law enforcement data; Step S42: performing time series analysis on the user's behavior record of retrieving law enforcement data to generate time series data of the retrieval behavior; Step S43: mining dynamic behavior features on the retrieval behavior time series data to generate dynamic retrieval behavior features; Step S44: Perform global portrait fitting on the dynamic retrieval behavior features to construct a user retrieval behavior portrait.

6. The method for managing data of a portable recorder based on on-site law enforcement according to claim 5, characterized in that: The specific steps of step S43 are: Extracting multiple call behavior timestamps based on the call behavior time series data; Calculate the frequency of the user's law enforcement data retrieval behavior records according to multiple retrieval behavior timestamps to obtain the retrieval behavior frequency; Perform data viewing time statistics on retrieved behavior time series data to obtain the viewing time of each section of law enforcement data; According to the frequency of retrieval behavior, the retrieval interval regularity analysis is performed on the viewing time of each section of law enforcement data to obtain the retrieval behavior regularity characteristic data; Identify the type of law enforcement data retrieved by users based on the retrieval behavior time series data; Performing a resource retrieval distribution analysis based on the type of law enforcement data retrieved by the user to obtain retrieval resource distribution data; Dynamic behavior feature analysis is performed on the resource distribution data and the regular feature data of the calling behavior to generate dynamic calling behavior features.

7. The method for managing data of a portable recorder based on on-site law enforcement according to claim 6, characterized in that: The specific steps of step S5 are: Step S51: Detect abnormal access behavior of the user access behavior profile and extract potential abnormal access behavior data; Step S52: Tracing the entire process of the user's retrieval of law enforcement data to obtain the entire retrieval process; Step S53: performing unauthorized access behavior analysis on potential abnormal access behavior data based on the full link of access behavior to obtain unauthorized access behavior data; Step S54: Performing a security risk quantitative assessment on the unauthorized access behavior data to generate an access behavior risk assessment value; Step S55: Make an adaptive permission diagnosis decision based on the retrieval behavior risk assessment value and build a data retrieval permission strategy.

8. The method for managing data of a portable recorder based on on-site law enforcement according to claim 7, characterized in that: The specific steps of step S6 are: Step S61: constructing a data tamper-proof architecture for a multi-level encrypted distributed storage space to generate a law enforcement data tamper-proof framework; Step S62: Mapping the authority decision rules of the law enforcement data tamper-proof framework based on the data access authority policy to construct the law enforcement data tamper-proof framework; Step S63: Perform dynamic access control on the law enforcement data authority tamper-proof framework and build a law enforcement data tamper-proof management model; The dynamic access control specifically includes: identifying the access permission level of the current user based on the data access permission policy; The access permission levels include: fully authorized user, partially authorized user and unauthorized user; When the user's access permission level is unauthorized user, the law enforcement data permission anti-tampering framework denies the user from accessing law enforcement data; When the user's access permission level is a partially authorized user, the law enforcement data permission anti-tampering framework allows access after desensitizing the law enforcement data; When the user's access permission level is fully authorized, the law enforcement data permission tamper-proof framework allows the user to fully access law enforcement data and record the chain of evidence of the access behavior.

9. A portable recorder data management system based on on-site law enforcement, which is used to implement the portable recorder data management method based on on-site law enforcement according to any one of claims 1 to 8, characterized in that: include: Data collection module, used to collect multi-dimensional law enforcement data from body recorders; Perform quality assessment and filtering on the multi-dimensional law enforcement data of the body recorder, and perform spatiotemporal label serialization processing to obtain a multi-dimensional law enforcement data sequence; An optimization module is used to adaptively compress and encode multi-dimensional law enforcement data sequences to generate multiple low-dimensional law enforcement feature tensors; perform end-to-end integrity verification optimization on multiple low-dimensional law enforcement feature tensors to generate optimized law enforcement low-dimensional tensors; The encrypted storage module is used to perform distributed storage mapping of the optimized law enforcement low-dimensional tensor on the blockchain to obtain a distributed storage space; perform multi-level encryption and obfuscation on the distributed storage space to obtain a multi-level encrypted distributed storage space; The portrait fitting module is used to obtain the user's behavior record of retrieving law enforcement data; perform dynamic behavior feature mining on the user's behavior record of retrieving law enforcement data, and construct a user's retrieval behavior portrait; The permission construction module is used to analyze the unauthorized access behavior of the user's access behavior profile, make adaptive permission diagnosis decisions, and build data access permission policies; The comprehensive management module performs dynamic access control on multi-level encrypted distributed storage space based on data retrieval permission strategies, and builds a tamper-proof management model for law enforcement data.

10. A body-worn recorder data management platform based on on-site law enforcement, the body-worn recorder data management platform based on on-site law enforcement comprising a memory and a processor, the memory storing computer-readable instructions, and when the computer-readable instructions are executed by the processor, the processor executes the body-worn recorder data management method based on on-site law enforcement as described in any one of claims 1-8.

Images