A monitoring and identification system based on a big data model
By combining multimodal data acquisition, multi-dimensional data modules, and big data platforms, and utilizing 5G smart camera arrays, infrared sensors, and deep learning models, the problem of insufficient data acquisition in complex environments by traditional monitoring systems has been solved. This enables timely early warning of potential dangers and effective identification of new types of illegal and criminal activities, thereby improving the efficiency of case investigation and the level of public safety management.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- 合肥徽通信息技术有限公司
- Filing Date
- 2025-07-17
- Publication Date
- 2026-05-15
AI Technical Summary
Traditional monitoring systems have shortcomings in integrating and processing diverse data, making it difficult to accurately identify behavioral characteristics in complex environments, unable to provide timely warnings of potential dangers, and unable to effectively respond to new types of illegal and criminal activities.
By employing a multimodal data acquisition module, a multi-source data import module, an algorithm module, and a big data platform, and combining technologies such as a 5G smart camera array, infrared sensors, and a voiceprint collector to form a multi-source heterogeneous data acquisition network, and using blockchain technology for encrypted transmission and storage, and deep learning models and convolutional neural networks for data processing, real-time acquisition of video, image, and sensor data has been achieved. The algorithm module, using blockchain technology for encrypted transmission and storage, and deep learning models and convolutional neural networks for data processing, combined with the data sharing service of the existing police cloud, provides data computing tools, enabling real-time acquisition of video, image, and sensor data. Data fusion through blockchain technology has constructed an intelligent data monitoring system.
It achieves comprehensiveness and timeliness over traditional monitoring systems, improves the timeliness and stability of information acquisition, enhances the ability to identify new types of illegal and criminal activities, and improves the efficiency of case investigation and the level of public safety management.
Smart Images
Figure CN120763818B_ABST
Abstract
Description
Technical Field
[0001] This application relates to the technical field of monitoring and identification systems based on big data models, and in particular to a monitoring and identification system based on big data models. Background Technology
[0002] In the current field of public safety management, the demand for surveillance and identification systems by public security departments is becoming increasingly urgent. Traditional surveillance systems mostly rely on single-modal data acquisition, such as acquiring video image information only through ordinary cameras. When facing complex and ever-changing real-world scenarios, the comprehensiveness and accuracy of information acquisition are severely insufficient. For example, in densely populated areas, it is difficult to effectively identify behavioral characteristics and provide early warnings of potential dangers.
[0003] Meanwhile, existing data processing technologies have significant shortcomings in integrating diverse data. This involves massive amounts of data, including resident population information, personnel information, hotel accommodation information, and personnel activity trajectories, but this data is often scattered across different systems, lacking efficient data fusion and analysis methods. Traditional algorithms have low accuracy in processing facial images and extracting and comparing human features, especially under complex conditions such as occlusion and changes in lighting, making it difficult to meet actual business needs.
[0004] Furthermore, with the continuous upgrading of new types of illegal and criminal methods, such as the increasingly covert and online posting of telecom fraud-related advertisements, traditional monitoring and identification systems are unable to analyze and process such abnormal behaviors in a timely and accurate manner, resulting in a lag in public safety management. Therefore, there is an urgent need for a monitoring and identification system that can integrate multimodal data, utilize advanced algorithms, and rely on a powerful data processing platform to improve the level of intelligence and the efficiency of public safety management. Summary of the Invention
[0005] To address the problems mentioned in the background section, this application provides a monitoring and identification system based on a big data model.
[0006] The monitoring and identification system based on a big data model provided in this application adopts the following technical solution:
[0007] A monitoring and identification system based on a big data model includes:
[0008] Multimodal data acquisition module: used for real-time acquisition of video, image and sensor data, consisting of a multi-source heterogeneous data acquisition network composed of a 5G smart camera array, infrared sensor and voiceprint collector;
[0009] Import multi-data module: Connect all resident population information, personnel information, hotel accommodation information, and activity trajectory data to the multimodal data acquisition module;
[0010] Establish an algorithm module: Build an algorithm repository to manage and model algorithms related to abnormal behavior analysis of posting fraudulent advertisements, as well as facial and human feature extraction and comparison algorithms.
[0011] Big Data Platform: Provides data computing tools for various business data analysis tasks and tag analysis, comprehensively improving the efficiency of business data analysis and model calculation;
[0012] Police Cloud: Based on the existing police cloud, it provides modeling-related view algorithm models, business data analysis models, and data aggregation and sharing services, as well as hardware support for front-end and back-end program operation, saving construction costs.
[0013] Preferably, the 5G smart camera array adopts a combination of wide-angle lens and telephoto lens. The wide-angle lens has a field of view of not less than 120°, and the telephoto lens supports optical zoom of not less than 30x, so as to realize large-scale scene monitoring and detail capture.
[0014] Preferably, the infrared sensor integrates a temperature sensing module, which can monitor the ambient temperature in real time and record abnormal temperature data to help determine potential dangerous situations.
[0015] Preferably, the imported multi-source data module uses blockchain technology to encrypt and transmit and store information on the city's permanent residents, personnel, hotel accommodations, and activity trajectories, ensuring data security and immutability.
[0016] Preferably, the algorithm for analyzing abnormal behavior of posting fraudulent advertisements uses a spatiotemporal trajectory clustering analysis method, combined with historical case location data, to predict high-risk posting areas and issue early warnings. The spatiotemporal trajectory clustering uses a density clustering algorithm, and its core calculation formula is: D(p,q) = where D(p,q) represents the spatiotemporal distance between trajectory points p and q, x and x' are the i-th spatial coordinates of points p and q respectively, and t and t' are the j-th time coordinates of points p and q respectively.
[0017] Preferably, the portrait and body feature extraction and comparison algorithm utilizes a deep learning model to extract and analyze the micro-expression and gait features of the portrait, thereby improving the recognition accuracy. Specifically, the micro-expression feature extraction employs a convolutional neural network (CNN), and the calculation formula for its convolutional layer is: x = f(x * m * k + b).
[0018] In the formula, x represents the j-th feature map of the l-th layer, x represents the i-th feature map of the (l-1)-th layer, k is the convolution kernel, b is the bias term, f is the activation function, and m is the set of input feature maps.
[0019] Preferably, the big data platform is equipped with a data hierarchical permission management module, which allocates corresponding data access and computing permissions according to the responsibilities of different business departments.
[0020] Preferably, the police cloud adopts a distributed storage architecture, which segments and stores the modeling-related view algorithm model and business data analysis model, and improves data access efficiency through load balancing technology.
[0021] Preferably, the multimodal data acquisition module and the multi-source data import module are connected via a high-speed optical fiber network with a transmission bandwidth of not less than 10Gbps to ensure the real-time performance and stability of data transmission.
[0022] In summary, at the data acquisition level, the multi-modal data acquisition module constructs a multi-source heterogeneous data acquisition network. Through the collaborative work of 5G smart camera arrays, infrared sensors, and voiceprint collectors, it achieves real-time acquisition of video, image, and sensor data. Compared to traditional single data acquisition methods, this significantly improves the comprehensiveness and timeliness of information acquisition. For example, the infrared sensor integrates a temperature sensing module, which can detect temperature anomalies in the early stages of dangerous events such as fires, buying valuable time for emergency response; the high-magnification zoom and wide-angle combination of the 5G smart camera can accurately identify distant targets and large-scale scene details, enhancing monitoring coverage capabilities.
[0023] The import of multi-source data modules integrates data from multiple sources, such as resident population data, with collected data, breaking down information silos. Based on blockchain technology, it ensures secure data transmission and storage, providing a richer and more accurate data foundation for public security operations. In case investigation, it allows for rapid correlation of various information, significantly shortening the investigation cycle.
[0024] An algorithm for analyzing abnormal behavior related to posting fraudulent advertisements targeting online telecom fraud was developed within the algorithm module. Utilizing spatiotemporal trajectory clustering formulas for intelligent analysis, it can predict high-risk areas by combining historical case data, enabling proactive early warning and effectively curbing new types of illegal and criminal activities. The facial and body feature extraction and comparison algorithms incorporate deep learning models and convolutional neural network formulas, significantly improving recognition accuracy. Even in complex environments, it can accurately locate targets, enhancing the efficiency of case investigation.
[0025] The big data platform provides professional data computing tools, optimizing business data analysis and model calculation processes, thus improving data analysis efficiency several times over. The police cloud provides hardware support based on existing network resources, employing a distributed storage architecture and load balancing technology, which not only reduces construction costs but also ensures efficient and stable system operation, providing solid technical support for public security operations. In summary, this system comprehensively enhances the level of intelligent management in public security and effectively safeguards public safety. Attached Figure Description
[0026] Figure 1 This is an overall block diagram of a monitoring and identification system based on a big data model, as described in an embodiment of this application. Detailed Implementation
[0027] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the embodiments of the present invention will be further described in detail below with reference to the accompanying drawings.
[0028] The illustrative embodiments and descriptions of the present invention are provided herein to explain the invention, but are not intended to limit the invention.
[0029] In this invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," "linking," and "fixing," etc., should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.
[0030] It should be understood that the terms "comprising / including," "consisting of," or any other variations are intended to cover non-exclusive inclusion, such that a product, apparatus, process, or method that comprises a list of elements includes not only those elements but may also include, where necessary, other elements not expressly listed, or elements inherent to such a product, apparatus, process, or method. Without further limitation, an element defined by the phrases "comprising / including," "consisting of," does not exclude the presence of additional identical elements in the product, apparatus, process, or method that includes said element.
[0031] It should also be understood that the terms "upper", "lower", "front", "rear", "left", "right", "top", "bottom", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device, component or structure referred to must have a specific orientation, be constructed or operated in a specific orientation, and should not be construed as a limitation of the present invention.
[0032] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.
[0033] like Figure 1 As shown, a monitoring and identification system based on a big data model includes:
[0034] Multimodal data acquisition module: used for real-time acquisition of video, image and sensor data, consisting of a multi-source heterogeneous data acquisition network composed of a 5G smart camera array, infrared sensor and voiceprint collector;
[0035] Import multi-data module: Connect all resident population information, personnel information, hotel accommodation information, and activity trajectory data to the multimodal data acquisition module;
[0036] Establish an algorithm module: Build an algorithm repository to manage and model algorithms related to abnormal behavior analysis of posting fraudulent advertisements, as well as facial and human feature extraction and comparison algorithms.
[0037] Big Data Platform: Provides data computing tools for various business data analysis tasks and tag analysis, comprehensively improving the efficiency of business data analysis and model calculation;
[0038] Police Cloud: Based on the existing police cloud, it provides modeling-related view algorithm models, business data analysis models, and data aggregation and sharing services, as well as hardware support for front-end and back-end program operation, saving construction costs.
[0039] In this embodiment, the 5G smart camera array adopts a combination of wide-angle lens and telephoto lens. The wide-angle lens has a field of view of not less than 120°, and the telephoto lens supports optical zoom of not less than 30x, so as to realize large-scale scene monitoring and detail capture.
[0040] In this embodiment, the infrared sensor integrates a temperature sensing module, which can monitor the ambient temperature in real time and record abnormal temperature data to help determine potential dangerous situations.
[0041] In this embodiment, the imported multi-source data module uses blockchain technology to encrypt and transmit and store information on the city's permanent residents, personnel, hotel accommodations, and activity trajectories, ensuring data security and immutability.
[0042] In this embodiment, the algorithm for analyzing abnormal behavior of posting fraudulent advertisements uses a spatiotemporal trajectory clustering analysis method, combined with historical crime scene data, to predict high-risk posting areas and issue early warnings. The spatiotemporal trajectory clustering uses a density clustering algorithm, and its core calculation formula is: D(p,q) = where D(p,q) represents the spatiotemporal distance between trajectory points p and q, x and x' are the i-th spatial coordinates of points p and q respectively, and t and t' are the j-th time coordinates of points p and q respectively.
[0043] In this embodiment, the portrait and human body feature extraction and comparison algorithm utilizes a deep learning model to extract and analyze the micro-expression and gait features of the portrait, thereby improving the recognition accuracy. Specifically, the micro-expression feature extraction employs a convolutional neural network (CNN), and the calculation formula for its convolutional layer is: x = f(x * m * k + b), where x represents the j-th feature map in the l-th layer, xi represents the i-th feature map in the (l-1)-th layer, k is the convolutional kernel, b is the bias term, f is the activation function, and m is the set of input feature maps.
[0044] In this embodiment, the big data platform is equipped with a data hierarchical permission management module, which allocates corresponding data access and calculation permissions according to the responsibilities of different business departments.
[0045] In this embodiment, the police cloud adopts a distributed storage architecture, which divides and stores the modeling-related view algorithm model and business data analysis model into shards, and improves data access efficiency through load balancing technology.
[0046] In this embodiment, the multimodal data acquisition module and the multi-source data import module are connected via a high-speed optical fiber network with a transmission bandwidth of not less than 10Gbps, ensuring the real-time performance and stability of data transmission.
[0047] The implementation principle of a monitoring and identification system based on a big data model, as described in this application embodiment, is as follows: First, the multimodal data acquisition module starts working. A multi-source heterogeneous data acquisition network composed of a 5G smart camera array, infrared sensors, and a voiceprint collector collects video, images, and sensor data in the monitored area in real time. The 5G smart camera uses a combination of wide-angle and telephoto lenses to capture high-definition images of a wide-area scene and transmits the collected video and image data in real time via the 5G network; the infrared sensor collects infrared image data while monitoring the ambient temperature; and the voiceprint collector collects sound data from the surrounding environment in real time. This multimodal data is synchronously transmitted to the system's data processing stage. Next, the multi-source data import module integrates resident population information, personnel information, hotel accommodation information, and activity trajectory data with the data acquired by the multimodal data acquisition module. During data transmission, blockchain technology is used to encrypt the data, ensuring the security and integrity of the data during transmission and storage. The integrated data forms a more comprehensive and richer dataset, providing a foundation for subsequent analysis. Subsequently, the algorithm module begins to function. The algorithm for analyzing abnormal behavior related to posting fraudulent advertisements is based on fused multi-source data. It uses spatiotemporal trajectory clustering formulas to analyze personnel activity trajectories and combines this with historical crime scene data to calculate the probability of fraudulent advertisement postings in each area. Once the probability exceeds a preset threshold, an early warning mechanism is triggered. The facial and body feature extraction and comparison algorithm uses deep learning models and convolutional neural network formulas to extract features from collected facial and body data and compare them with personnel data in the system database to identify potential targets. The big data platform receives the data results processed by the algorithm modules and, for various public security business data analysis tasks and tag analysis, uses professional data computing tools to further mine and analyze the data, generating visualized analysis reports and tags to provide data support for public security decision-making. Finally, the police cloud provides hardware support for the entire system's operation. Its distributed storage architecture segments and stores modeling-related view algorithm models and business data analysis models, and uses load balancing technology to rationally allocate computing resources, ensuring the efficient and stable operation of the front-end and back-end programs of each module. Meanwhile, the police cloud also provides data aggregation and sharing services, facilitating data interaction and collaborative work among various public security departments, achieving information sharing, and improving the overall processing efficiency of public security operations.
[0048] Finally, the following points should be noted: First, in the description of this application, it should be noted that, unless otherwise specified and limited, the terms "installation", "connection", and "linkage" should be interpreted broadly, and can be mechanical or electrical connections, or internal connections between two components, or direct connections. "Up", "down", "left", "right", etc. are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may change.
[0049] Secondly: The accompanying drawings of the embodiments disclosed in this invention only involve the structures involved in the embodiments disclosed in this invention. Other structures can refer to the general design. In the absence of conflict, the same embodiment and different embodiments of this invention can be combined with each other.
[0050] In conclusion, 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, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.
Claims
1. A monitoring and identification system based on a big data model, characterized in that: include: Multimodal data acquisition module: used for real-time acquisition of video, image and sensor data, consisting of a multi-source heterogeneous data acquisition network composed of a 5G smart camera array, infrared sensor and voiceprint collector; Import multi-data module: Connect all resident population information, personnel information, hotel accommodation information, and activity trajectory data to the multimodal data acquisition module; Algorithm module establishment: An algorithm repository is established to manage the abnormal behavior analysis algorithm for posting fraudulent advertisements, the facial and human feature extraction and comparison algorithm involved in modeling. The abnormal behavior analysis algorithm for posting fraudulent advertisements adopts a spatiotemporal trajectory clustering analysis method, combined with historical case location data, to predict high-risk posting areas and issue early warnings. Among them, the spatiotemporal trajectory clustering adopts a density clustering algorithm. The facial and human feature extraction and comparison algorithm uses a deep learning model to extract and analyze the micro-expression and gait features of the facial image to improve the recognition accuracy. Among them, the micro-expression feature extraction adopts a convolutional neural network. Big Data Platform: Provides data computing tools for various business data analysis tasks and tag analysis, comprehensively improving the efficiency of business data analysis and model calculation; Police Cloud: Based on the existing police cloud, it provides modeling-related view algorithm models, business data analysis models, and data aggregation and sharing services, as well as hardware support for front-end and back-end program operation, saving construction costs.
2. The monitoring and identification system based on a big data model according to claim 1, characterized in that, The 5G smart camera array uses a combination of wide-angle and telephoto lenses. The wide-angle lens has a field of view of no less than 120°, and the telephoto lens supports optical zoom of no less than 30x, enabling large-scale scene monitoring and detail capture.
3. The monitoring and identification system based on a big data model according to claim 1, characterized in that, The infrared sensor integrates a temperature sensing module, which can monitor the ambient temperature in real time and record abnormal temperature data to help identify potential dangerous situations.
4. The monitoring and identification system based on a big data model according to claim 1, characterized in that, The imported multi-source data module uses blockchain technology to encrypt and transmit and store information on the city's permanent residents, personnel, hotel accommodations, and activity trajectories, ensuring data security and immutability.
5. A monitoring and identification system based on a big data model according to claim 1, characterized in that, The big data platform is equipped with a data hierarchical permission management module, which allocates corresponding data access and computing permissions according to the responsibilities of different business departments.
6. The monitoring and identification system based on a big data model according to claim 1, characterized in that, The police cloud adopts a distributed storage architecture, which divides and stores the modeling-related view algorithm model and business data analysis model into shards, and improves data access efficiency through load balancing technology.
7. The monitoring and identification system based on a big data model according to claim 1, characterized in that, The multimodal data acquisition module and the multi-source data import module are connected via a high-speed fiber optic network with a transmission bandwidth of no less than 10Gbps, ensuring the real-time performance and stability of data transmission.