Behavior monitoring system with low resource consumption

Abstract

The invention relates to a behavior monitoring system with low resource consumption. The behavior monitoring system with low resource consumption comprises a data acquisition module which is used for acquiring video data around to-be-protected equipment based on preset video acquisition equipment; the data processing module is used for analyzing the video data based on a preset sliding window judgment mechanism and extracting a to-be-identified key frame; the behavior recognition module is used for performing behavior recognition on the key frame based on a preset visual target detection algorithm; and the user interface module is used for monitoring real-time data and receiving system alarms. Through real-time video monitoring, violation events can be identified and recorded, prevention measures such as screen locking can be automatically taken, and the information safety level is effectively improved. Meanwhile, by introducing a background blurring algorithm, the privacy of irrelevant personnel or objects in the scene is further protected, and only key information is ensured to be processed and recorded.

Description

Technical Field

[0001] The present disclosure relates to the technical field of information security and computer monitoring, and in particular to a behavior monitoring system with low resource consumption. Background Art

[0002] In the field of information security, it is extremely important to protect sensitive data from unauthorized access and leakage. With the popularization of high-resolution camera equipment, the risk of illegally shooting screen content has greatly increased. Traditional protective measures such as anti-peeping film and signal interference are no longer sufficient to fully deal with this problem. At the same time, conventional user behavior monitoring systems require users to manually lock the screen when leaving their workstations. This measure that relies on user initiative often fails due to negligence.

[0003] In addition, existing behavior monitoring solutions, especially those based on deep learning, often require high computing power and large memory, which leads to their reliance on external processing chips. This not only increases costs, but also limits the popularity and flexibility of the system. For example, these systems cannot run directly on ordinary laptops or desktop computers, but need to rely on additional hardware support.

[0004] Therefore, one or more methods are needed to solve the above problems.

[0005] It should be noted that the information disclosed in the above background technology section is only used to enhance the understanding of the background of the present disclosure, and therefore may include information that does not constitute the prior art known to ordinary technicians in the field. Summary of the invention

[0006] The purpose of the present disclosure is to provide a behavior monitoring system with low resource consumption, thereby overcoming one or more problems caused by limitations and defects of related technologies at least to a certain extent.

[0007] According to one aspect of the present disclosure, a behavior monitoring system with low resource consumption is provided, comprising:

[0008] A data acquisition module, the data acquisition module is used to collect video data around the device to be protected based on a preset video acquisition device;

[0009] A data processing module, the data processing module is used to analyze the video data based on a preset sliding window judgment mechanism and extract key frames to be identified;

[0010] A behavior recognition module, the behavior recognition module is used to perform behavior recognition on the key frame based on a preset visual target detection algorithm;

[0011] A user interface module is used to monitor real-time data and receive system alarms.

[0012] In an exemplary embodiment of the present disclosure, the data acquisition module of the system further includes:

[0013] The data acquisition module is used to acquire video data around the device to be protected based on a built-in or external camera that establishes a data connection with the device to be protected.

[0014] In an exemplary embodiment of the present disclosure, the preset sliding window judgment mechanism in the data processing module of the system further includes:

[0015] The first detection frequency is preset as the initial detection frequency, and the system operates at the first detection frequency;

[0016] If the behavior identified within the preset time window is a potential violation, the system switches to a second detection frequency, and the frequency value of the second detection frequency is greater than the first detection frequency.

[0017] In an exemplary embodiment of the present disclosure, the system further includes:

[0018] After the system switches to the second detection frequency, if the behavior recognition does not identify any preset potential violation within a preset number of consecutive time windows, the system will switch to the first detection frequency.

[0019] In an exemplary embodiment of the present disclosure, the preset visual target detection algorithm in the behavior recognition module in the system further includes:

[0020] The behavior recognition of the key frames is performed based on a preset improved version of the YOLO target detection model as a visual target detection algorithm.

[0021] In an exemplary embodiment of the present disclosure, the behavior recognition module of the system further includes:

[0022] In the behavior recognition module, key targets are recognized on the key frames based on a preset visual target detection algorithm, and behavior recognition is completed based on the key targets;

[0023] After the behavior recognition module performs behavior recognition on the key frame based on a preset visual target detection algorithm, Gaussian blur processing is performed on the background area of ​​the key target.

[0024] In an exemplary embodiment of the present disclosure, the system further includes:

[0025] Based on the behavior recognition result, an emergency response index is calculated, and the emergency response index is compared with a preset safety threshold, and an alarm signal is generated based on the comparison result.

[0026] In one aspect of the present disclosure, a behavior monitoring method with low resource consumption is provided, comprising:

[0027] Collect video data around the device to be protected based on a preset video acquisition device;

[0028] Analyzing the video data based on a preset sliding window judgment mechanism to extract key frames to be identified;

[0029] Performing behavior recognition on the key frames based on a preset visual target detection algorithm;

[0030] An alarm signal is generated based on the behavior recognition result, and the screen is automatically locked and the administrator is notified based on the alarm signal.

[0031] A low-resource-consuming behavior monitoring system in an exemplary embodiment of the present disclosure includes a data acquisition module, which is used to collect video data around the device to be protected based on a preset video acquisition device; a data processing module, which is used to analyze the video data based on a preset sliding window judgment mechanism and extract key frames to be identified; a behavior recognition module, which is used to perform behavior recognition on the key frames based on a preset visual target detection algorithm; and a user interface module, which is used to monitor real-time data and receive system alarms. The present disclosure significantly reduces the demand for hardware resources for behavior monitoring by optimizing memory management and data processing based on a sliding window judgment mechanism.

[0032] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] The above and other features and advantages of the present disclosure will become more apparent by describing in detail example embodiments thereof with reference to the attached drawings.

[0034] Figure 1 A schematic block diagram of a low resource consumption behavior monitoring system according to an exemplary embodiment of the present disclosure is shown;

[0035] Figure 2 A module schematic diagram of a low resource consumption behavior monitoring system according to an exemplary embodiment of the present disclosure is shown;

[0036] Figure 3 A flow chart of a behavior monitoring method with low resource consumption according to an exemplary embodiment of the present disclosure is shown. DETAILED DESCRIPTION

[0037] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in many forms and should not be construed as limited to the embodiments set forth herein; rather, these embodiments are provided so that this disclosure will be comprehensive and complete and will fully convey the concepts of the example embodiments to those skilled in the art. The same reference numerals in the figures represent the same or similar parts, and thus their repeated description will be omitted.

[0038] In addition, the described features, structures or characteristics may be combined in one or more embodiments in any suitable manner. In the following description, many specific details are provided to provide a full understanding of the embodiments of the present disclosure. However, those skilled in the art will appreciate that the technical solutions of the present disclosure may be practiced without one or more of the specific details, or other methods, components, materials, devices, steps, etc. may be adopted. In other cases, known structures, methods, devices, implementations, materials or operations are not shown or described in detail to avoid blurring the various aspects of the present disclosure.

[0039] The block diagrams shown in the accompanying drawings are merely functional entities and do not necessarily correspond to physically independent entities. That is, these functional entities may be implemented in software form, or these functional entities or parts of functional entities may be implemented in one or more software hardened modules, or these functional entities may be implemented in different networks and / or processor devices and / or microcontroller devices.

[0040] In this exemplary embodiment, a low resource consumption behavior monitoring system 100 is first provided; Figure 1 As shown in FIG. 1 , the low resource consumption behavior monitoring system 100 includes:

[0041] A data acquisition module 110, wherein the data acquisition module 110 is used to acquire video data around the device to be protected based on a preset video acquisition device;

[0042] A data processing module 120, the data processing module 120 is used to analyze the video data based on a preset sliding window judgment mechanism and extract key frames to be identified;

[0043] A behavior recognition module 130, wherein the behavior recognition module 130 is used to perform behavior recognition on the key frame based on a preset visual target detection algorithm;

[0044] The user interface module 140 is used to monitor real-time data and receive system alarms.

[0045] A low-resource-consuming behavior monitoring system in an exemplary embodiment of the present disclosure includes a data acquisition module, which is used to collect video data around the device to be protected based on a preset video acquisition device; a data processing module, which is used to analyze the video data based on a preset sliding window judgment mechanism and extract key frames to be identified; a behavior recognition module, which is used to perform behavior recognition on the key frames based on a preset visual target detection algorithm; and a user interface module, which is used to monitor real-time data and receive system alarms. The present disclosure significantly reduces the demand for hardware resources for behavior monitoring by optimizing memory management and data processing based on a sliding window judgment mechanism.

[0046] Next, a low resource consumption behavior monitoring system 100 in this example embodiment will be further described.

[0047] The low resource consumption behavior monitoring system 100 of the present disclosure includes:

[0048] The data acquisition module 110 is used to acquire video data around the device to be protected based on a preset video acquisition device.

[0049] In the embodiment of this example, the data acquisition module 110 of the system further includes:

[0050] The data acquisition module 110 is used to acquire video data around the device to be protected based on a built-in or external camera that establishes a data connection with the device to be protected.

[0051] The data processing module 120 is used to analyze the video data based on a preset sliding window judgment mechanism and extract key frames to be identified.

[0052] In the embodiment of this example, the preset sliding window judgment mechanism in the data processing module 120 of the system further includes:

[0053] The first detection frequency is preset as the initial detection frequency, and the system operates at the first detection frequency;

[0054] If the behavior identified within the preset time window is a potential violation, the system switches to a second detection frequency, and the frequency value of the second detection frequency is greater than the first detection frequency.

[0055] In the embodiment of this example, the system further comprises:

[0056] After the system switches to the second detection frequency, if the behavior recognition does not identify any preset potential violation within a preset number of consecutive time windows, the system will switch to the first detection frequency.

[0057] The behavior recognition module 130 is used to perform behavior recognition on the key frame based on a preset visual target detection algorithm.

[0058] In the embodiment of this example, the preset visual target detection algorithm in the behavior recognition module 130 in the system further includes:

[0059] The behavior recognition of the key frames is performed based on a preset improved version of the YOLO target detection model as a visual target detection algorithm.

[0060] In the embodiment of this example, the behavior recognition module 130 of the system further includes:

[0061] The behavior recognition module 130 performs key target recognition on the key frame based on a preset visual target detection algorithm, and completes behavior recognition based on the key target;

[0062] After the behavior recognition module 130 performs behavior recognition on the key frame based on a preset visual target detection algorithm, Gaussian blur processing is performed on the background area of ​​the key target.

[0063] The user interface module 140 is used to monitor real-time data and receive system alarms.

[0064] In the embodiment of this example, the system further comprises:

[0065] Based on the behavior recognition result, an emergency response index is calculated, and the emergency response index is compared with a preset safety threshold, and an alarm signal is generated based on the comparison result.

[0066] In this exemplary embodiment, if Figure 2 As shown, the system implementation and deployment of the present disclosure includes:

[0067] (1) System architecture

[0068] The architecture of this system is designed as four main components: data acquisition module 110, data processing module 120, behavior recognition module 130, and user interface module 140. The data acquisition module 110 collects video data in real time through a built-in or external camera. The data processing module 120 applies a sliding window judgment mechanism to optimize processing performance and selects key frames for further analysis. The behavior recognition module 130 uses the trained YOLO model to detect violations and identify key targets, and uses the OpenVINO toolkit to optimize the model's reasoning performance on devices without graphics cards. The user interface module 140 provides a visualization platform for system administrators to monitor real-time data and receive system alerts.

[0069] (2) Hardware and software requirements

[0070] The system can run without relying on high-end hardware and is suitable for standard laptops and desktop computers without independent graphics cards. The basic hardware requirements of the system include a processor with at least 4GB of memory and a standard 1080p camera. In terms of software, the system is developed based on C++ and uses the OpenVINO toolkit to optimize the inference process of the YOLO model to improve operating efficiency and reduce hardware requirements. The system supports deployment on Windows 7 and Window 10, 11.

[0071] (3) Usage scenarios and practical applications

[0072] The system is designed to accommodate a variety of surveillance environments, including corporate offices, educational institutions, and public areas. In corporate offices, the system can prevent sensitive information from being leaked through illegal means such as cell phone photography. In educational environments, it can be used to monitor and prevent cheating in exams. In public places such as libraries or conference rooms, the system helps enhance personal privacy protection and prevent illegal use of personal devices. Administrators can customize surveillance parameters through the user interface, such as adjusting detection frequency and blur depth, to suit different surveillance needs and environments.

[0073] It should be noted that, although several modules or units of a behavior monitoring system 100 with low resource consumption are mentioned in the above detailed description, such division is not mandatory. In fact, according to an embodiment of the present disclosure, the features and functions of two or more modules or units described above can be embodied in one module or unit. Conversely, the features and functions of one module or unit described above can be further divided into multiple modules or units to be embodied.

[0074] Furthermore, in this exemplary embodiment, if Figure 3 As shown, the present disclosure also provides a behavior monitoring method with low resource consumption. The behavior monitoring method with low resource consumption may include:

[0075] Step S110: collecting video data around the device to be protected based on a preset video collection device.

[0076] Step S120: analyzing the video data based on a preset sliding window judgment mechanism to extract key frames to be identified.

[0077] Step S130: performing behavior recognition on the key frame based on a preset visual target detection algorithm.

[0078] Step S140, generating an alarm signal based on the behavior recognition result, and automatically locking the screen based on the alarm signal and notifying the administrator.

[0079] In the embodiment of this example, the low-resource consumption behavior monitoring system 100 of the present disclosure is intended to automatically identify and prevent violations such as using mobile phones, cameras and other devices to illegally shoot screen contents, and monitor the situation where the user does not perform a lock screen operation when leaving the workstation. The present disclosure combines image recognition, behavior analysis and machine learning technologies, and through real-time video monitoring, it can not only identify and record violations, but also automatically take preventive measures such as locking the screen, effectively improving the level of information security. At the same time, by introducing a background blur algorithm, the privacy of unrelated people or objects in the scene is further protected to ensure that only critical information is processed and recorded. Especially in environments that require high security and privacy protection, such as financial services, research and development institutions, and public safety, the present invention provides an effective and economical solution to prevent the leakage of sensitive information and maintain the safety of operating sites.

[0080] In the embodiment of this example, the automatic screen lock and illegal shooting prevention mechanism includes:

[0081] (1) Real-time video monitoring

[0082] The system uses real-time video streaming from built-in or external cameras for monitoring, and continuously analyzes video data through deep learning models to detect whether any illegal devices (such as mobile phones or cameras) appear in the monitored area.

[0083] (2) Automatic identification of illegal filming behavior

[0084] The present invention uses a modified version of the YOLO object detection model, which is specially optimized for fast identification of small electronic devices. The YOLO model divides the image into multiple regions at once and predicts the object category and its bounding box in each region. The specific formula is as follows:

[0085]

[0086] p(class|object) represents the probability that an object belongs to a certain class when an object is detected, and p(object) represents the probability that an object exists in any region. Represents the intersection-over-union ratio between the predicted box and the true box. In this way, the model can quickly and accurately detect electronic devices.

[0087] (3) Automatic lock screen response

[0088] Once the model identifies illegal filming behavior, the system will calculate an emergency response index E and compare it with the set safety threshold T:

[0089] E=αΔF(t)+βp 违规 (t)

[0090] Where α and β are weight coefficients that adjust the difference value and the violation probability p 违规 Contribution to the response index. If E ≥ T, the system will automatically lock the screen and notify the administrator.

[0091] The sliding window judgment mechanism includes:

[0092] (1) Dynamic detection frequency adjustment principle

[0093] The core of the present invention is to use a time window (time sliding window) to dynamically adjust the frequency of violation detection. This method is based on an observation: in monitoring scenarios, violations are usually sudden and infrequent. Therefore, using constant high-frequency detection not only consumes a lot of hardware resources, but may also cause unnecessary processing delays.

[0094] By setting an initial detection frequency f0, the system runs at a frequency of f0 in normal times, ensuring low resource consumption. When a potential violation is detected within any given time window T, the system will automatically increase the detection frequency to f0. hign , in order to more intensively monitor and confirm the persistence of violations.

[0095] Detection frequency adjustment strategy

[0096] The adjustment of detection frequency follows the following formula:

[0097]

[0098] f(t) represents the detection frequency within time t, and T is the length of the time window considered. This strategy ensures a quick response after a violation occurs while minimizing resource usage when there is no violation.

[0099] (2) Confirmation of Violations and Frequency Reset

[0100] Once the system increases the detection frequency, it will detect the next n time windows T with f high If no violation is detected again in these consecutive windows, the system will determine that the violation has stopped and the detection frequency will be reset to f0, thereby reducing resource usage.

[0101] No violation is detected within n consecutive windows: f(t+nT)=f0

[0102] Through this method of dynamically adjusting the detection frequency, the present invention not only improves the response speed and accuracy to violations, but also significantly reduces the system's demand for hardware resources, especially in an environment with low violation frequency, which can greatly improve the efficiency and practicality of the system.

[0103] In the embodiment of this example, target recognition and selective blurring include:

[0104] The present invention uses the target detection model to accurately identify key targets in the image, such as faces, mobile phones, and cameras. In this process, the system only processes these defined targets instead of performing full image segmentation, which significantly reduces the computing resources and time required for processing.

[0105] In the blur implementation method, once key targets are identified, the system will apply Gaussian blur to the background areas around these targets. This method ensures that the images of faces and key devices remain clear, while other information in the background, such as documents or screens that may contain sensitive data, is effectively blurred, thereby reducing the risk of being illegally photographed. The specific implementation of Gaussian blur is as follows:

[0106]

[0107] Among them, B(x, y) represents the original background pixel value, B′(x, y) is the blurred background pixel value, σ is the Gaussian blur standard deviation that controls the degree of blur, and W is the normalization factor to ensure that the sum of the processed pixel values ​​remains unchanged.

[0108] In the embodiment of this example, the low resource consumption behavior monitoring system 100 disclosed in the present invention can achieve the following beneficial effects: Technological innovation, using YOLO and OpenVINO optimization: By integrating the YOLO target detection model with Intel's OpenVINO toolkit, the system can run efficiently on ordinary computers that do not have high-end GPU support. This optimization makes model reasoning faster while maintaining high recognition accuracy, allowing even devices with limited resources to perform complex image analysis tasks; Dynamic sliding window judgment mechanism: The time sliding window technology is used to adjust the detection frequency, so that the system increases the detection frequency when potential violations are detected, and reduces the frequency when no violations are detected, effectively balancing monitoring efficiency and resource consumption; Resource efficiency, reducing hardware requirements: Through OpenVINO optimization, the system reduces dependence on graphics cards, allowing users to deploy the monitoring system without investing in expensive graphics processing hardware. This is particularly suitable for small businesses and educational institutions with limited budgets or inconvenient equipment updates. Reduce memory and processing resource consumption: The dynamically adjusted sliding window mechanism automatically adjusts the data processing load according to the real-time monitoring needs, reduces invalid processing, saves memory usage, and extends the service life of the device; User convenience, simplified system deployment and maintenance: The system is designed with user ease of use in mind, supports rapid deployment on multiple operating systems, and has a user-friendly interface, which is convenient for non-professional users to perform daily maintenance and adjust monitoring settings; Flexible application scenarios, the system can be widely used in a variety of environments, such as offices, schools, and public spaces, and administrators can adjust monitoring settings such as detection frequency and blur depth according to different environmental requirements; Privacy protection, efficient background blur technology: By applying Gaussian blur only to the background around key targets, this system protects the privacy of irrelevant people and objects in the scene. This selective blur ensures that sensitive information such as files or other personal data behind the screen is not leaked.

[0109] Data processing in compliance with privacy regulations: The system is designed to comply with the latest data protection regulations, such as GDPR, to ensure that all monitoring activities and data processing activities are performed within the scope permitted by law.

[0110] It should be noted that, although the steps of the method in the present disclosure are described in a specific order in the drawings, this does not require or imply that the steps must be performed in this specific order, or that all the steps shown must be performed to achieve the desired results. Additionally or alternatively, some steps may be omitted, multiple steps may be combined into one step, and / or one step may be decomposed into multiple steps, etc.

[0111] In addition, the above-mentioned figures are only schematic illustrations of the processes included in the method according to an exemplary embodiment of the present invention, and are not intended to be limiting. It is easy to understand that the processes shown in the above-mentioned figures do not indicate or limit the time sequence of these processes. In addition, it is also easy to understand that these processes can be performed synchronously or asynchronously, for example, in multiple modules.

[0112] Those skilled in the art will readily appreciate other embodiments of the present disclosure after considering the specification and practicing the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the present disclosure that follow the general principles of the present disclosure and include common knowledge or customary technical means in the art that are not disclosed in the present disclosure. The specification and examples are to be considered as exemplary only, and the true scope and spirit of the present disclosure are indicated by the claims.

[0113] It should be understood that the present disclosure is not limited to the exact structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.

Claims

1. A low resource consumption behavior monitoring system, characterized in that: The system comprises: A data acquisition module, the data acquisition module is used to collect video data around the device to be protected based on a preset video acquisition device; A data processing module, the data processing module is used to analyze the video data based on a preset sliding window judgment mechanism and extract key frames to be identified; A behavior recognition module, the behavior recognition module is used to perform behavior recognition on the key frame based on a preset visual target detection algorithm; A user interface module is used to monitor real-time data and receive system alarms.

2. The system according to claim 1, characterized in that The data acquisition module of the system also includes: The data acquisition module is used to acquire video data around the device to be protected based on a built-in or external camera that establishes a data connection with the device to be protected.

3. The system according to claim 1, characterized in that The preset sliding window judgment mechanism in the data processing module of the system also includes: The first detection frequency is preset as the initial detection frequency, and the system operates at the first detection frequency; If the behavior identified within the preset time window is a potential violation, the system switches to a second detection frequency, and the frequency value of the second detection frequency is greater than the first detection frequency.

4. The system according to claim 3, characterized in that The system further comprises: After the system switches to the second detection frequency, if the behavior recognition does not identify any preset potential violation within a preset number of consecutive time windows, the system will switch to the first detection frequency.

5. The system according to claim 1, wherein: The preset visual target detection algorithm in the behavior recognition module of the system also includes: The behavior recognition of the key frames is performed based on a preset improved version of the YOLO target detection model as a visual target detection algorithm.

6. The system according to claim 5, characterized in that The behavior recognition module of the system also includes: In the behavior recognition module, key targets are recognized on the key frames based on a preset visual target detection algorithm, and behavior recognition is completed based on the key targets; After the behavior recognition module performs behavior recognition on the key frame based on a preset visual target detection algorithm, Gaussian blur processing is performed on the background area of ​​the key target.

7. The system according to claim 6, characterized in that The system further comprises: Based on the behavior recognition result, an emergency response index is calculated, and the emergency response index is compared with a preset safety threshold, and an alarm signal is generated based on the comparison result.

8. A behavior monitoring method with low resource consumption, characterized in that: The method comprises: Collect video data around the device to be protected based on a preset video acquisition device; Analyzing the video data based on a preset sliding window judgment mechanism to extract key frames to be identified; Performing behavior recognition on the key frames based on a preset visual target detection algorithm; An alarm signal is generated based on the behavior recognition result, and the screen is automatically locked and the administrator is notified based on the alarm signal.

9. An electronic device, characterized in that: include Processor; and A memory having computer-readable instructions stored thereon, wherein the computer-readable instructions implement the method according to claim 8 when executed by the processor.

10. A computer-readable storage medium, characterized in that: A computer program is stored thereon, which implements the method according to claim 8 when executed by a processor.

Images