Data security management method and system of network camera

Abstract

The invention is suitable for the field of camera data security, and provides a data security management method and system for a network camera, and the system comprises an audio and video information collection module, a motion mode recognition module, a motion state analysis module, an audio feature analysis module, and a result integration and alarm module. The motion vector of the object between different frames can be calculated in real time by accurately tracking the motion track of the object in the video, so that the coherence of the motion mode is effectively identified. Through the deep analysis, any unnatural motion jump or interruption can be quickly identified, and potential editing or tampering behaviors can be found in time. Through analysis of audio waveforms and frequency spectrums, the method can ensure high consistency of audio content and video content. Integrity and synchronism analysis of the audio can reveal whether the audio conforms to behaviors in the video, and help to judge whether audio missing or improper editing exists.

Description

Technical Field

[0002] The invention belongs to the field of video data security, and in particular relates to a data security management method and system for a network camera. Background Art

[0003] The data security of network cameras mainly involves protecting the video and audio data collected by the cameras to prevent unauthorized access, data tampering and privacy leakage. These cameras are usually connected to the Internet and transmit data over the network, exposing them to multiple security threats such as hacker attacks and data interception.

[0004] To ensure data security, first of all, strong authentication and access control measures need to be implemented to ensure that only authorized users can access the camera and its data. In addition, encryption technology for video streams and stored data is also an important security measure that can effectively prevent data from being stolen or tampered with during transmission and storage. In addition, regularly updating the camera's firmware and software can patch potential security vulnerabilities, thereby improving the overall security of the system.

[0005] Traditional surveillance systems often rely only on video recordings, but lack in-depth analysis and verification of video content. This makes it possible for tampering to occur without any alarm, which greatly reduces the credibility of surveillance data. Surveillance videos may be maliciously tampered with or edited, especially when key events occur, some personnel may delete videos of specific time periods to cover up unfavorable evidence. This behavior not only affects the integrity of the video, but may also lead to misunderstandings or misjudgments of the incident, which in turn affects the fairness of investigations and legal proceedings. Summary of the invention

[0006] The purpose of the present invention is to provide a data security management method for a network camera, aiming to solve the technical problems existing in the prior art identified in the background technology.

[0007] The present invention is implemented in this way: a data security management method for a network camera, the method comprising: Get the video data continuously recorded by the network camera and record the audio data synchronously; Extract data frames from video data, collect and calculate motion vectors between frames in the video, and identify and record the motion patterns of objects; Count the movement frequency and speed of each object, analyze whether there are jumps or interruptions in the identified movement trajectory, and determine and mark whether the movement in the video is natural and coherent; Acquire audio data synchronized with video data, extract audio waveform and spectrum data, analyze the characteristics of audio signals, analyze the time domain characteristics of audio and video respectively, and check the correspondence between audio waveform and video frame; Integrate the results of video analysis and audio and video analysis to determine whether the video is complete and coherent.

[0008] As a further solution of the present invention, the extracting of data frames in the video data, collecting and calculating motion vectors between frames in the video, and identifying and recording the motion pattern of the object specifically include: Extract each frame from the continuously recorded video data and pre-process it; Detect and record feature points in each frame, calculate the displacement of each feature point in adjacent frames, and calculate the motion vector between adjacent frames; Initialize the motion trajectory data structure for each identified object. In each frame, update the object's motion trajectory according to the calculated motion vector and record the motion trajectory of each object into the data structure, including timestamp, displacement and motion direction. Motion features are extracted from the motion trajectory, and the motion pattern of the object is classified based on the motion features.

[0009] As a further solution of the present invention, the statistics of the movement frequency and speed of each object, the analysis of whether the identified movement trajectory has jumps or interruptions, and the determination and marking of whether the movement in the video is natural and coherent specifically include: Analyze the displacement and time difference between adjacent points in the motion trajectory, identify the speed changes of objects in different time periods, and calculate the average speed of each object; Set the jump threshold of the motion trajectory, analyze whether the object's motion trajectory is continuous, detect whether there are incoherent changes, and if so, mark the video frames with incoherent changes; The statistical frequency and speed are compared with the normal behavior pattern to determine whether it meets expectations, and the movement trajectories that are significantly different from the normal pattern are identified and marked.

[0010] As a further solution of the present invention, the analysis of the characteristics of the audio signal, respectively analyzing the time domain characteristics of the audio and video, and checking the correspondence between the audio waveform and the video frame specifically includes: Synchronize with the video data and extract the audio data with the same timestamp; Perform FFT on the audio signal, convert it to the frequency domain, and visualize the spectrum results; The audio waveform is fully aligned with the timestamps of the video frames, and the synchronization of the audio waveform and the action in the video is analyzed frame by frame; Identify portions of the audio data that are inconsistent with the video data and mark them.

[0011] As a further solution of the present invention, the results of integrating the video analysis and the audio and video analysis to determine whether the video is complete and coherent specifically include: Summarize the motion trajectory analysis results and audio feature analysis results to form a data set; Analyze the motion trajectory analysis results and the audio feature analysis results respectively to see whether there are marked parts; If there is a marked portion, the video data or audio data corresponding to the mark of the portion is extracted in a truncated manner and feedback information is generated.

[0012] Another object of the present invention is to provide a data security management system for network cameras, the system comprising: The audio and video information acquisition module is used to obtain the video data continuously recorded by the network camera and simultaneously record the audio data; The motion pattern recognition module is used to extract data frames from the video data, collect and calculate the motion vectors between frames in the video, and recognize and record the motion pattern of the object; The motion state analysis module is used to count the motion frequency and speed of each object, analyze whether there are jumps or interruptions in the identified motion trajectory, and determine and mark whether the motion in the video is natural and coherent; The audio feature analysis module is used to obtain audio data synchronized with the video data, extract audio waveform and spectrum data, analyze the characteristics of the audio signal, analyze the time domain characteristics of the audio and video respectively, and check the corresponding relationship between the audio waveform and the video frame; The result integration and alarm module is used to integrate the results of video analysis and audio and video analysis to determine whether the video is complete and coherent.

[0013] As a further solution of the present invention, the motion pattern recognition module includes: A video frame extraction and preprocessing unit, used to extract each frame from the continuously recorded video data and preprocess it; A feature point detection and displacement calculation unit, used to detect and record feature points in each frame, calculate the displacement of each feature point in adjacent frames, and calculate the motion vector between adjacent frames; A motion trajectory initialization and update unit, which is used to initialize the motion trajectory data structure for each identified object, update the object's motion trajectory according to the calculated motion vector in each frame, and record the motion trajectory of each object into the data structure, including timestamp, displacement and motion direction; The motion feature extraction and classification unit is used to extract motion features from the motion trajectory and classify the motion mode of the object based on the motion features.

[0014] As a further solution of the present invention, the motion state analysis module includes: The speed analysis and calculation unit is used to analyze the displacement and time difference between adjacent points in the motion trajectory, identify the speed changes of objects in different time periods, and calculate the average speed of each object; A motion continuity detection unit is used to set a jump threshold of the motion trajectory, analyze whether the motion trajectory of the object is continuous, detect whether there are incoherent changes, and if so, mark the video frames with incoherent changes; The behavior pattern comparison and anomaly identification unit is used to compare the statistical frequency and speed with the normal behavior pattern to determine whether it meets expectations, and to identify and mark the movement trajectory that is significantly different from the normal pattern.

[0015] As a further solution of the present invention, the audio feature analysis module includes: An audio data synchronization extraction unit, used to synchronize with the video data and extract audio data with the same time stamp; The frequency domain conversion and visualization unit is used to perform FFT on the audio signal, convert it to the frequency domain, and visualize the spectrum result; An audio and video time alignment analysis unit, which is used to completely align the timestamps of the audio waveform with the video frames, and analyze the synchronization of the audio waveform with the action in the video frame by frame; The inconsistent part identification and marking unit is used to identify the part of the audio data that is inconsistent with the video data and mark it.

[0016] As a further solution of the present invention, the result integration and alarm module includes: A data set aggregation unit, used to aggregate the motion trajectory analysis results and the audio feature analysis results to form a data set; A marked part analysis unit, used to analyze whether there is a marked part in the motion trajectory analysis result and the audio feature analysis result respectively; The data truncation extraction and feedback generation unit is used to perform truncation extraction on the video data or audio data corresponding to the mark of the part if there is a marked part, and generate feedback information.

[0017] The beneficial effects of the present invention are: By accurately tracking the motion trajectory of objects in the video, the motion vector of the object between different frames can be calculated in real time, effectively identifying the consistency of the motion pattern. This deep analysis allows any unnatural motion jumps or interruptions to be quickly identified, and potential editing or tampering can be discovered in a timely manner.

[0018] At the same time, by analyzing the audio waveform and spectrum, the method can ensure a high degree of consistency between the audio content and the video content. The integrity and synchronization analysis of the audio can reveal whether the audio matches the behavior in the video and help determine whether there is audio missing or improper editing.

[0019] This multi-dimensional analysis not only improves the security of video surveillance, but also enhances the traceability of data, ensuring that reliable evidence can be provided when faced with disputes or incident investigations. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 A flow chart of a data security management method for a network camera provided by an embodiment of the present invention; Figure 2 A flowchart of extracting data frames from video data, collecting and calculating motion vectors between frames in the video, and identifying and recording the motion pattern of an object provided by an embodiment of the present invention; Figure 3 A flowchart for performing statistics on the movement frequency and speed of each object, analyzing whether the identified movement trajectory has jumps or interruptions, and determining and marking whether the movement in the video is natural and coherent, provided in an embodiment of the present invention; Figure 4 A flowchart of analyzing the characteristics of an audio signal, analyzing the time domain characteristics of audio and video respectively, and checking the correspondence between audio waveforms and video frames provided by an embodiment of the present invention; Figure 5 A flowchart for integrating the results of video analysis and audio and video analysis to determine whether a video is complete and coherent, provided by an embodiment of the present invention; Figure 6 A structural diagram of a data security management system for a network camera provided by an embodiment of the present invention; Figure 7 A structural block diagram of a motion pattern recognition module package provided by an embodiment of the present invention; Figure 8 A structural block diagram of a motion state analysis module provided by an embodiment of the present invention; Fig. 9 A structural block diagram of an audio feature analysis module provided in an embodiment of the present invention; Fig.10 This is a structural block diagram of the result integration and alarm module provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0021] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0022] It is understood that the terms "first", "second", etc. used in this application may be used herein to describe various elements, but unless otherwise specified, these elements are not limited by these terms. These terms are only used to distinguish a first element from another element. For example, without departing from the scope of this application, a first xx script may be referred to as a second xx script, and similarly, a second xx script may be referred to as a first xx script.

[0023] Figure 1 The flowchart of the data security management method of the network camera provided by the embodiment of the present invention is as follows: Figure 1 As shown, the method includes: S100, obtaining video data continuously recorded by the network camera and synchronously recording audio data; In this step, the continuously recorded video data stream is extracted from the network camera. Usually, the real-time streaming protocol (RTSP) or other suitable protocols can be used to ensure the real-time and stability of data acquisition. Next, the video stream is decomposed into individual frames, usually extracted at a rate of 30 frames per second. During this process, an appropriate frame rate is applied to ensure the accuracy of motion analysis. Each frame image needs to be preprocessed, including denoising, histogram equalization, scaling and other operations to improve the accuracy and efficiency of subsequent feature detection. Denoising can use methods such as Gaussian blur or median filtering to ensure the stability of feature points.

[0024] S200, extracting data frames from the video data, collecting and calculating motion vectors between frames in the video, and identifying and recording the motion pattern of the object; This step first extracts each frame from the continuously recorded video data. This process usually involves breaking up the video stream according to a set frame rate (for example, 30 frames per second) so that each frame can be analyzed independently. After each frame is extracted, it must first be preprocessed to improve the effectiveness of subsequent analysis. Preprocessing steps include denoising, which reduces random noise in the image through techniques such as Gaussian blurring or median filtering to improve the detection accuracy of feature points. In addition, image enhancement, such as histogram equalization, can be performed to enhance the contrast of the image and thus improve the recognizability of feature points.

[0025] Next, feature points are detected and recorded in each frame. Common feature point detection algorithms include Harris corner detection, Shi-Tomasi or SIFT. These feature points are important reference points in the video and can provide stable tracking information. For each feature point, its displacement in adjacent frames is calculated, and the corresponding feature point in adjacent frames is found through the feature matching algorithm, and finally the motion vector between adjacent frames is obtained. Each motion vector represents the direction and speed of the object's movement between frames.

[0026] Initializing the data structure of the motion trajectory for each identified object is a key step in S2. This structure will store the motion trajectory of each object, including timestamp, displacement, and direction of motion. In each frame, the object's motion trajectory is updated based on the calculated motion vector, and this information is recorded in the data structure to ensure the continuity and accuracy of the object's motion. By analyzing these motion trajectories, motion features such as speed, acceleration, motion pattern, etc. can be extracted. These features will help classify the object's motion pattern for subsequent abnormal behavior detection.

[0027] Through the careful processing of each frame and the precise detection of feature points, the calculation results of the motion vector are ensured to be highly reliable, which is crucial for subsequent motion pattern recognition and anomaly detection. In addition, the dynamic update and feature extraction of the motion trajectory enable the system to track the behavior of objects in real time and promptly identify any unnatural or suspicious motion patterns. This feature is particularly important in real-time monitoring and security management. Through in-depth analysis of motion features, the system can better adapt to changes in different environments, enhance the robustness of the algorithm, and ensure that it can still maintain good performance in complex scenarios such as lighting changes and object occlusion.

[0028] like Figure 2 The extracting of data frames from the video data, collecting and calculating motion vectors between frames in the video, and identifying and recording the motion pattern of the object specifically include: S210, extracting each frame from the continuously recorded video data and preprocessing the frame; S220, detecting and recording feature points in each frame, calculating the displacement of each feature point in adjacent frames, and calculating the motion vector between adjacent frames; S230, initializing a motion trajectory data structure for each identified object, updating the motion trajectory of the object in each frame according to the calculated motion vector, and recording the motion trajectory of each object into the data structure, including a timestamp, a displacement, and a motion direction; S240, extracting motion features from the motion trajectory, and classifying the motion pattern of the object based on the motion features.

[0029] S300, counting the movement frequency and speed of each object, analyzing whether the recognized movement trajectory has jumps or interruptions, and determining and marking whether the movement in the video is natural and coherent; In this step, the displacement and time difference between adjacent points in the motion trajectory are analyzed, and these parameters are calculated to identify the speed changes of the object in different time periods. For example, if the speed of an object suddenly increases or decreases over a period of time, it may indicate that the object's behavior is abnormal.

[0030] Next, the system calculates the average speed of each object in order to establish a baseline value as a reference for subsequent analysis. By setting the jump threshold of the motion trajectory, the system can effectively distinguish between normal movement and unnatural changes. For example, if the motion trajectory of an object suddenly jumps sharply in a certain frame, it will be marked as suspicious. At this time, the system will mark the video frames with incoherent changes to provide a basis for subsequent manual review or automatic alarm.

[0031] In addition, based on the statistical frequency and speed, the system will also compare the object's movement with normal behavior patterns. By learning and establishing common movement patterns, the system can determine whether the currently identified movement trajectory meets expectations. If an object's movement frequency or speed is found to deviate significantly from the normal pattern, the system will mark the movement trajectory to draw attention.

[0032] Through detailed analysis of movement trajectories, potential suspicious behaviors can be identified in a timely manner, which is especially important in security monitoring and event review. In addition, the mechanism of setting jump thresholds enhances the sensitivity of the system, enabling it to effectively identify small incoherent changes, rather than just obvious anomalies. This meticulous analysis method not only improves the reliability of monitoring data, but also provides a solid foundation for subsequent event investigations.

[0033] like Figure 3 The above-mentioned method of counting the movement frequency and speed of each object, analyzing whether there are jumps or interruptions in the identified movement trajectory, and judging and marking whether the movement in the video is natural and coherent, specifically includes: S310, analyzing the displacement and time difference between adjacent points in the motion trajectory, identifying the speed change of the object in different time periods, and calculating the average speed of each object; S320, setting a jump threshold of the motion trajectory, analyzing whether the motion trajectory of the object is continuous, detecting whether there is an incoherent change, and if so, marking the video frame with the incoherent change; S330, comparing the statistical frequency and speed with the normal behavior pattern to determine whether it meets expectations, and identifying and marking the movement trajectory that is significantly different from the normal pattern.

[0034] S400, acquiring audio data synchronized with the video data, extracting audio waveform and spectrum data, analyzing characteristics of the audio signal, analyzing time domain characteristics of the audio and video respectively, and checking the corresponding relationship between the audio waveform and the video frame; In this step, audio data with the same timestamp is extracted from the synchronously recorded video data. This process ensures that the audio data is completely aligned with the video data in time, laying the foundation for subsequent analysis. Next, the extracted audio signal is subjected to a fast Fourier transform (FFT) to convert it from the time domain to the frequency domain. Frequency domain analysis can reveal the spectral characteristics of the audio signal, making it easier to identify the intensity and changes of different frequency components, and visualize the spectral results. The distribution, intensity and change trend of the frequency components are displayed through charts, allowing analysts to intuitively capture the characteristics of the audio signal.

[0035] At the same time, the system will fully align the timestamps of the audio waveform and the video frame to ensure that each frame of video strictly corresponds to the corresponding audio clip, providing an accurate data basis for frame-by-frame analysis. The synchronization of the audio waveform and the action in the video is analyzed frame by frame, and the changes in the audio signal when the corresponding action occurs in the video are checked, for example, whether the detected sound is consistent with the movement, interaction or event of the object in the video. By comparing the audio waveform with the video data, the system can identify the parts of the audio data that are inconsistent with the video data, such as the lack of sound accompanying a certain action or the mismatch between the audio signal and the action, and mark the inconsistent parts found for subsequent review and analysis. These marks may point to potential editing, forgery or other anomalies.

[0036] By ensuring accurate synchronization between audio and video through frame-by-frame analysis, which is crucial for surveillance and security applications, potential forgery or cropping can be revealed in a timely manner. In addition, by combining the time domain and frequency domain feature analysis of audio, not only the movement of objects in the video is monitored, but also contextual information is provided for the background sound or dialogue of the event, which enhances the richness of data and the depth of analysis. Real-time audio and video comparative analysis can quickly capture unnatural events, such as the mismatch between the sound and the action in the video, and provide timely alarms so that necessary measures can be taken. By visualizing the spectrum results, analysts can intuitively understand the characteristics of the audio signal, allowing non-professionals to make judgments through visual charts. Finally, marking and recording inconsistent parts provides a clear basis for subsequent detailed reports and incident investigations, improving the efficiency and accuracy of incident handling.

[0037] like Figure 4 The analysis of the characteristics of the audio signal, respectively analyzing the time domain characteristics of the audio and video, and checking the correspondence between the audio waveform and the video frame specifically include: S410, extracting audio data with the same time stamp in synchronization with the video data; S420, performing FFT on the audio signal, converting it into the frequency domain, and visualizing the spectrum result; S430, completely aligning the timestamps of the audio waveform and the video frame, and analyzing the synchronization of the audio waveform and the action in the video frame by frame; S440, identifying the part of the audio data that is inconsistent with the video data and marking it.

[0038] S500 integrates the results of the video analysis and the audio and video analysis to determine whether the video is complete and coherent.

[0039] This step aggregates the motion trajectory analysis results with the audio feature analysis results to form a comprehensive data set. This process not only makes the information more centralized, but also provides a clear basis for subsequent analysis. By combining motion trajectories with audio features, the system is able to evaluate the video from multiple dimensions to identify potential anomalies or inconsistencies.

[0040] Next, the system will analyze the motion trajectory analysis results and audio feature analysis results to see if there are marked parts. These marked parts usually refer to anomalies detected in the previous steps, such as sudden changes in motion trajectories or asynchrony between audio and video. When the system finds these marked parts, it will perform a truncated extraction of the video data or audio data corresponding to the part. This process can not only accurately identify potential problems, but also provide intuitive evidence for detailed reports.

[0041] Finally, the generated feedback information will include a detailed description of the suspicious time period and a detailed description of the abnormal situation. This feedback information is a summary of the analysis results, which helps analysts quickly understand and evaluate the nature of the incident and take necessary follow-up measures.

[0042] By summarizing the analysis results of motion trajectories and audio features, the system can comprehensively evaluate the quality and credibility of the video from multiple angles. This multi-dimensional analysis improves the sensitivity of anomaly detection and can promptly identify and report potential problems. In addition, the truncated extraction function allows analysts to directly access the data of suspicious parts, facilitating detailed investigation and verification. The generated feedback information provides a clear basis for reporting and subsequent processing, making the entire data management process more efficient and flexible.

[0043] like Figure 5 The results of the integrated video analysis and audio and video analysis are shown to determine whether the video is complete and coherent, specifically including: S510, summarizing the motion trajectory analysis results and the audio feature analysis results to form a data set; S520, analyzing the motion trajectory analysis result and the audio feature analysis result respectively to see whether there is a marked part; S530: If there is a marked portion, perform truncation extraction on the video data or audio data corresponding to the mark of the portion, and generate feedback information.

[0044] Figure 6The structural block diagram of the data security management system of the network camera provided by the embodiment of the present invention is as follows: Figure 6 As shown, the system comprises: The audio and video information acquisition module 100 is used to obtain the video data continuously recorded by the network camera and simultaneously record the audio data; The motion pattern recognition module 200 is used to extract data frames from the video data, collect and calculate the motion vectors between the frames in the video, and recognize and record the motion pattern of the object; The motion state analysis module 300 is used to collect statistics on the motion frequency and speed of each object, analyze whether there are jumps or interruptions in the identified motion trajectory, and determine and mark whether the motion in the video is natural and coherent; The audio feature analysis module 400 is used to obtain audio data synchronized with the video data, extract audio waveform and spectrum data, analyze the characteristics of the audio signal, analyze the time domain characteristics of the audio and video respectively, and check the corresponding relationship between the audio waveform and the video frame; The result integration and alarm module 500 is used to integrate the results of video analysis and audio and video analysis to determine whether the video is complete and coherent.

[0045] like Figure 7 The motion pattern recognition module 200 shown includes: The video frame extraction and preprocessing unit 210 is used to extract each frame from the continuously recorded video data and preprocess it; A feature point detection and displacement calculation unit 220, used to detect and record feature points in each frame, calculate the displacement of each feature point in adjacent frames, and calculate the motion vector between adjacent frames; A motion trajectory initialization and update unit 230 is used to initialize a motion trajectory data structure for each identified object, update the motion trajectory of the object according to the calculated motion vector in each frame, and record the motion trajectory of each object into the data structure, including a timestamp, displacement, and motion direction; The motion feature extraction and classification unit 240 is used to extract motion features from the motion trajectory and classify the motion mode of the object based on the motion features.

[0046] like Figure 8 The motion state analysis module 300 shown includes: The speed analysis and calculation unit 310 is used to analyze the displacement and time difference between adjacent points in the motion trajectory, identify the speed change of the object in different time periods, and calculate the average speed of each object; A motion continuity detection unit 320 is used to set a jump threshold of the motion trajectory, analyze whether the motion trajectory of the object is continuous, detect whether there is an incoherent change, and if so, mark the video frame with the incoherent change; The behavior pattern comparison and abnormality identification unit 330 is used to compare the statistical frequency and speed with the normal behavior pattern to determine whether it meets expectations, and to identify and mark the movement trajectory that is significantly different from the normal pattern.

[0047] like Fig. 9 The audio feature analysis module 400 shown includes: An audio data synchronization extraction unit 410, used to synchronize with the video data and extract audio data with the same time stamp; The frequency domain conversion and visualization unit 420 is used to perform FFT on the audio signal, convert it to the frequency domain, and visualize the spectrum result; An audio and video time alignment analysis unit 430, for completely aligning the timestamps of the audio waveform with the video frames, and analyzing the synchronization of the audio waveform with the action in the video frame by frame; The inconsistent part identification and marking unit 440 is used to identify the part of the audio data that is inconsistent with the video data and mark it.

[0048] like Fig.10 The result integration and alarm module 500 shown includes: A data set aggregation unit 510, used to aggregate the motion trajectory analysis results and the audio feature analysis results to form a data set; A marked part analysis unit 520, used to analyze whether there is a marked part in the motion trajectory analysis result and the audio feature analysis result respectively; The data truncation extraction and feedback generation unit 530 is used to perform truncation extraction on the video data or audio data corresponding to the mark of the part if there is a marked part, and generate feedback information.

[0049] It should be understood that, although each step in the flow chart of each embodiment of the present invention is shown in sequence according to the indication of the arrow, these steps are not necessarily performed in sequence according to the order indicated by the arrow. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be performed in other orders. Moreover, at least a portion of the steps in each embodiment may include a plurality of sub-steps or a plurality of stages, and these sub-steps or stages are not necessarily performed at the same time, but can be performed at different times, and the execution order of these sub-steps or stages is not necessarily performed in sequence, but can be performed in turn or alternately with at least a portion of other steps or sub-steps or stages of other steps. Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the program can be stored in a non-volatile computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0050] The technical features of the above-described embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above-described embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0051] The above-mentioned embodiments only express several implementation methods of the present invention, and the description thereof is relatively specific and detailed, but it cannot be understood as limiting the scope of the patent of the present invention. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, which all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

[0052] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A data security management method for a network camera, characterized in that: The method comprises: Get the video data continuously recorded by the network camera and record the audio data synchronously; Extract data frames from video data, collect and calculate motion vectors between frames in the video, and identify and record the motion patterns of objects; Count the movement frequency and speed of each object, analyze whether there are jumps or interruptions in the identified movement trajectory, and determine and mark whether the movement in the video is natural and coherent; Acquire audio data synchronized with video data, extract audio waveform and spectrum data, analyze the characteristics of audio signals, analyze the time domain characteristics of audio and video respectively, and check the correspondence between audio waveform and video frame; Integrate the results of video analysis and audio and video analysis to determine whether the video is complete and coherent.

2. The method according to claim 1, characterized in that The extracting of data frames in the video data, collecting and calculating motion vectors between frames in the video, and identifying and recording the motion pattern of the object specifically include: Extract each frame from the continuously recorded video data and pre-process it; Detect and record feature points in each frame, calculate the displacement of each feature point in adjacent frames, and calculate the motion vector between adjacent frames; Initialize the motion trajectory data structure for each identified object. In each frame, update the object's motion trajectory according to the calculated motion vector and record the motion trajectory of each object into the data structure, including timestamp, displacement and motion direction. Motion features are extracted from the motion trajectory, and the motion pattern of the object is classified based on the motion features.

3. The method according to claim 2, characterized in that The statistics of the movement frequency and speed of each object, the analysis of whether the identified movement trajectory has jumps or interruptions, and the determination and marking of whether the movement in the video is natural and coherent specifically include: Analyze the displacement and time difference between adjacent points in the motion trajectory, identify the speed changes of objects in different time periods, and calculate the average speed of each object; Set the jump threshold of the motion trajectory, analyze whether the object's motion trajectory is continuous, detect whether there are incoherent changes, and if so, mark the video frames with incoherent changes; The statistical frequency and speed are compared with the normal behavior pattern to determine whether it meets expectations, and the movement trajectories that are significantly different from the normal pattern are identified and marked.

4. The method according to claim 3, characterized in that The step of analyzing the characteristics of the audio signal, analyzing the time domain characteristics of the audio and video respectively, and checking the correspondence between the audio waveform and the video frame specifically includes: Synchronize with the video data and extract the audio data with the same timestamp; Perform FFT on the audio signal, convert it to the frequency domain, and visualize the spectrum results; The audio waveform is fully aligned with the timestamps of the video frames, and the synchronization of the audio waveform and the action in the video is analyzed frame by frame; Identify portions of the audio data that are inconsistent with the video data and mark them.

5. The method according to claim 4, characterized in that The results of integrating video analysis and audio and video analysis to determine whether the video is complete and coherent specifically include: Summarize the motion trajectory analysis results and audio feature analysis results to form a data set; Analyze the motion trajectory analysis results and the audio feature analysis results respectively to see whether there are marked parts; If there is a marked portion, the video data or audio data corresponding to the mark of the portion is extracted in a truncated manner and feedback information is generated.

6. The data security management system of the network camera is characterized by: The system comprises: The audio and video information acquisition module is used to obtain the video data continuously recorded by the network camera and simultaneously record the audio data; The motion pattern recognition module is used to extract data frames from the video data, collect and calculate the motion vectors between frames in the video, and recognize and record the motion pattern of the object; The motion state analysis module is used to count the motion frequency and speed of each object, analyze whether there are jumps or interruptions in the identified motion trajectory, and determine and mark whether the motion in the video is natural and coherent; The audio feature analysis module is used to obtain audio data synchronized with the video data, extract audio waveform and spectrum data, analyze the characteristics of the audio signal, analyze the time domain characteristics of the audio and video respectively, and check the corresponding relationship between the audio waveform and the video frame; The result integration and alarm module is used to integrate the results of video analysis and audio and video analysis to determine whether the video is complete and coherent.

7. The system according to claim 6, characterized in that The motion pattern recognition module comprises: A video frame extraction and preprocessing unit, used to extract each frame from the continuously recorded video data and preprocess it; A feature point detection and displacement calculation unit, used to detect and record feature points in each frame, calculate the displacement of each feature point in adjacent frames, and calculate the motion vector between adjacent frames; A motion trajectory initialization and update unit, which is used to initialize the motion trajectory data structure for each identified object, update the object's motion trajectory according to the calculated motion vector in each frame, and record the motion trajectory of each object into the data structure, including timestamp, displacement and motion direction; The motion feature extraction and classification unit is used to extract motion features from the motion trajectory and classify the motion mode of the object based on the motion features.

8. The system according to claim 7, characterized in that The motion state analysis module comprises: The speed analysis and calculation unit is used to analyze the displacement and time difference between adjacent points in the motion trajectory, identify the speed changes of objects in different time periods, and calculate the average speed of each object; A motion continuity detection unit is used to set a jump threshold of the motion trajectory, analyze whether the motion trajectory of the object is continuous, detect whether there are incoherent changes, and if so, mark the video frames with incoherent changes; The behavior pattern comparison and anomaly identification unit is used to compare the statistical frequency and speed with the normal behavior pattern to determine whether it meets expectations, and to identify and mark the movement trajectory that is significantly different from the normal pattern.

9. The system according to claim 8, characterized in that The audio feature analysis module comprises: An audio data synchronization extraction unit, used to synchronize with the video data and extract audio data with the same time stamp; The frequency domain conversion and visualization unit is used to perform FFT on the audio signal, convert it to the frequency domain, and visualize the spectrum result; An audio and video time alignment analysis unit, which is used to completely align the timestamps of the audio waveform with the video frames, and analyze the synchronization of the audio waveform with the action in the video frame by frame; The inconsistent part identification and marking unit is used to identify the part of the audio data that is inconsistent with the video data and mark it.

10. The system according to claim 9, characterized in that The result integration and alarm module includes: A data set aggregation unit, used to aggregate the motion trajectory analysis results and the audio feature analysis results to form a data set; A marked part analysis unit, used to analyze whether there is a marked part in the motion trajectory analysis result and the audio feature analysis result respectively; The data truncation extraction and feedback generation unit is used to perform truncation extraction on the video data or audio data corresponding to the mark of the part if there is a marked part, and generate feedback information.

Images