Camera monitoring system based on wireless communication and communication protocol optimization method

Abstract

The invention discloses a camera monitoring system based on wireless communication and a communication protocol optimization method, and relates to the technical field of communication optimization, the camera monitoring system is composed of a monitoring camera group, a wireless communication network, an edge computing unit, a cloud data center and a user interaction terminal; the monitoring camera group simulates human eye vision, adjusts a visual angle and a resolution ratio according to needs, and collects data; a wireless communication network is combined with quantum encryption and a traditional technology, and a communication mode is intelligently selected; the edge calculation unit processes data with low power consumption by using a neuromorphic chip; the cloud data center fuses the block chain and federal learning analysis data; the user interaction terminal has holographic projection and emotion perception functions and provides personalized services. According to the invention, the modules are collaboratively innovated, the pertinence of collected data is strong, the communication is efficient and safe, the processing is fast and accurate, and the interaction experience is good; the system has the functions of intelligent analysis, energy-saving management and the like, the intelligence and reliability of the system are improved, and the system has wide application prospects and market value.

Description

Technical Field

[0001] The present invention relates to the technical field of monitoring communication optimization, and in particular to a camera monitoring system based on wireless communication and a communication protocol optimization method. Background Art

[0002] In today's society, the demand for security and monitoring is growing. As an important means to ensure public safety, maintain social order, and protect personal and property safety, camera monitoring systems are widely used in various fields. However, the existing camera monitoring systems have many shortcomings and are difficult to meet the actual needs of continuous development.

[0003] Most traditional camera surveillance systems use wired connections, which brings many limitations in practical applications. The wiring process is not only cumbersome and complicated, but also requires a lot of manpower, material resources and time costs, and will cause a certain degree of damage to the original structure of the building. In addition, the wired connection method greatly limits the installation location of the camera, lacks flexibility, and is difficult to adjust and expand in time according to actual monitoring needs. When it is necessary to add monitoring points in a new area, it is often necessary to re-lay the cables, which undoubtedly increases the construction and maintenance costs of the system.

[0004] In terms of data transmission, existing wireless communication technologies also have some problems when applied to camera monitoring systems. Common wireless communication methods such as Wi-Fi have limited transmission distances and weak anti-interference capabilities. In complex environments, they are prone to signal instability and data loss, resulting in freezing and discontinuity of the monitoring screen, and are unable to accurately reflect the actual situation of the monitoring area in real time. Moreover, with the widespread use of high-definition cameras, the amount of monitoring data has increased dramatically, which has put forward higher requirements for the bandwidth and speed of data transmission. Existing wireless communication technologies are difficult to meet the fast and stable transmission of such large-scale data.

[0005] In terms of data processing and analysis, traditional camera monitoring systems mainly rely on simple video recording and storage, and lack the ability to deeply mine and analyze monitoring data. A large amount of monitoring video data is stored, but the valuable information contained in it has not been fully utilized. For example, it is impossible to conduct real-time analysis and early warning of human behavior and object movement trajectories in the monitoring area, making it difficult to detect abnormal situations in time and take corresponding measures, which greatly reduces the effectiveness of the monitoring system.

[0006] In terms of system security, the existing camera monitoring system also has certain hidden dangers. Due to the imperfect encryption technology in the data transmission and storage process, the monitoring data is easily stolen and tampered with, resulting in privacy leaks and security accidents. In addition, the system lacks effective fault diagnosis and self-repair capabilities. Once a hardware failure or software vulnerability occurs, manual troubleshooting and repair are often required, which not only consumes time and energy, but may also affect the normal operation of the monitoring system.

[0007] With the continuous development of emerging technologies such as the Internet of Things, artificial intelligence, and quantum communication, new opportunities have been provided for the upgrade and improvement of camera monitoring systems. Therefore, the development of a camera monitoring system based on wireless communication and a communication protocol optimization method has important practical significance, which can effectively solve the problems existing in the existing monitoring system and promote the development of monitoring technology towards intelligence, efficiency, and security. Summary of the invention

[0008] The present invention proposes a camera monitoring system based on wireless communication and a communication protocol optimization method to solve the problems mentioned in the above-mentioned prior art.

[0009] In order to achieve the above object, the present invention adopts the following technical solution: a camera monitoring system based on wireless communication, comprising the following modules: Surveillance camera group: A camera array that uses the bionic vision principle to simulate the visual perception mechanism of the human eye and uses an adaptive visual attention model Dynamically adjust the shooting angle and resolution, including is the i-th visual feature, is the weight of the corresponding feature, n is the visual collection sample; Wireless communication network: A hybrid communication mode combining quantum encryption communication technology and traditional wireless communication technology is introduced; during data transmission, according to the real-time signal strength S, bandwidth requirement B, transmission distance D and security level requirement L, a comprehensive evaluation formula is used to evaluate the Select the best communication method, where is the weight coefficient; Edge computing unit: Equipped with an edge computing platform based on neuromorphic computing chips, it processes data by imitating the working mode of human brain neurons; by building a feature extraction model Complete the feature extraction task, where is the jth pulse feature, is the weight of the corresponding feature; Cloud data center: Adopts a distributed blockchain and federated learning architecture, uses deep reinforcement transfer learning algorithm to analyze data, and uses reward function to Optimize model parameters, where is the reward value at the tth time step; User interaction terminal: Through holographic projection technology, emotion recognition sensors and natural language processing technology, it can perceive the user's emotional state and language intention, and use personalized recommendation models Automatically adjust the interactive interface and provide personalized services, including is the user's preference feature, is the weight of the corresponding feature.

[0010] Furthermore, the following modules are also included: Intelligent behavior analysis module: Based on the model of the fusion of spatiotemporal graph convolutional network ST-GCN and generative adversarial network GAN, it analyzes the target behavior trajectory and action mode in the monitoring area; Represent the spatiotemporal relationship of the target and use ST-GCN to extract spatiotemporal features ; The generator G in GAN attempts to generate behavior patterns, and the discriminator D distinguishes between real and generated behavior patterns. Through the behavior feature vector V and time series T, the formula Calculate the probability P of target behavior abnormality and issue an early warning when P exceeds the preset threshold. is the weight coefficient and b is the bias term.

[0011] Energy management module: A composite power supply method combining solar panels, rechargeable batteries and wireless energy collection technology; by monitoring the remaining battery power , Solar panel power generation , Wireless Energy Harvesting Power and system power consumption , using the formula Calculate the battery life T; when T is lower than the set value, the system uses the energy allocation model to allocate energy based on device priority and task importance. Automatically adjust the system working mode, including is the energy consumption requirement of the ith device, is the priority weight of the corresponding device.

[0012] Security encryption module: adopts an encryption method that combines quantum key distribution QKD with homomorphic encryption; QKD is used to generate and distribute quantum keys, and homomorphic encryption allows data processing in the ciphertext state. At the data sending end, the data is first encrypted once with the quantum key generated by QKD, and then the encrypted data is encrypted again using the homomorphic encryption algorithm; the receiving end performs the corresponding reverse decryption operation; the quantum key is further obfuscated by introducing a chaotic sequence encryption algorithm to enhance the security of the key. The encryption formula is: ,in is the original quantum key, C is the chaotic sequence, and f is the chaotic encryption function.

[0013] Adaptive image enhancement module: The module combines the Retinex algorithm with the generative adversarial network GAN to enhance the image; according to the ambient light intensity L, image clarity index C, contrast index K and image noise level N, the formula ),in This is the image processed by the Retinex algorithm. Optimized images generated by GAN, is the weight coefficient, and Dynamically adjust camera image acquisition parameters and evaluate image quality through image quality indicators ,in is the weight coefficient.

[0014] Multi-sensor fusion module: Integrates temperature, humidity, smoke, motion, and gas concentration sensors, and introduces a sensor data fusion algorithm based on a Bayesian network; the algorithm builds a Bayesian network model The data from different sensors are analyzed and fused, where G is a directed acyclic graph, indicating the dependency relationship between sensors. is the conditional probability distribution), according to the sensor data , using the Bayesian formula Calculate the probability of different events, where H is the hypothetical event, is the prior probability of the hypothesized event, is the probability of sensor data occurring when the hypothetical event occurs, is the probability of sensor data occurrence.

[0015] System self-diagnosis and optimization module: Adopt PHM technology based on fault prediction and health management; regularly test the system hardware performance, software operation status and communication quality, build a deep convolutional autoencoder DCAE model to extract features and detect anomalies in the system operation data, and compare the feature vectors under normal conditions. and the feature vector at the current state , using the similarity measurement formula Calculate the similarity S; when S is lower than the set threshold, determine that the system may have a fault; based on the detection result R and the historical optimization record H, use the reinforcement learning algorithm to generate the optimization strategy O, and use the reward function Automatically adjust system parameters to improve overall performance, including is the reinforcement learning reward at the i-th time step.

[0016] A communication protocol optimization method for a camera monitoring system based on wireless communication is provided, comprising the following steps: Data collection steps: The surveillance camera group uses bionic vision principles and adaptive visual attention models to dynamically adjust the shooting angle and resolution according to the importance and motion status of the monitored target, and collect audio and video data in the monitored area; Communication selection step: The wireless communication network uses a comprehensive evaluation formula to select a communication method that combines quantum encryption communication technology with traditional wireless communication technology based on real-time signal strength S, bandwidth requirement B, transmission distance D, and security level requirement L to transmit data collected by the surveillance camera group; Edge processing steps: The edge computing unit uses a feature extraction model based on a neuromorphic computing chip and a spiking neural network (SNN) algorithm to preliminarily process the collected data and extract key features; Cloud analysis step: The cloud data center uses a distributed blockchain and federated learning architecture to analyze the received data using a reinforcement transfer learning algorithm and reward function to provide support for monitoring decisions; Result feedback step: The user interaction terminal uses holographic projection interaction and emotion perception functions to perceive the user's emotional state and language intention based on the personalized recommendation model, and feeds back the analysis results to the user through the cloud data center.

[0017] Furthermore, the method further comprises the following steps: Security encryption steps: The security encryption module uses an encryption method that combines quantum key distribution QKD with homomorphic encryption, and uses a chaotic sequence encryption algorithm to confuse the quantum key to ensure the security of data transmission and storage; Image enhancement step: The adaptive image enhancement module combines the Retinex algorithm with the generative adversarial network GAN to dynamically adjust the camera image acquisition parameters through formulas and image quality evaluation indicators according to the ambient light intensity L, image clarity index C, contrast index K, and image noise level N to improve image quality; Behavior analysis steps: The intelligent behavior analysis module is based on a model that combines the spatiotemporal graph convolutional network ST-GCN with the generative adversarial network GAN. It uses a formula to analyze the probability of abnormal target behavior and issues an early warning when the probability exceeds the threshold.

[0018] Energy management steps: The energy management module uses a composite power supply method that combines solar panels, rechargeable batteries and wireless energy collection technology to calculate the battery life T according to the formula; when T is lower than the set value, the system working mode is automatically adjusted through the energy allocation model to reduce overall energy consumption; Data fusion step: sensor fusion module integration, using the sensor data fusion algorithm based on Bayesian network and Bayesian formula to perform correlation analysis and fusion on the data of different sensors to generate environmental information and monitoring warning; System optimization steps: The system self-diagnosis and optimization module adopts PHM technology based on system fault prediction and health management, detects system faults through similarity measurement formula, and uses learning algorithms and reward functions to generate and execute optimization strategies.

[0019] Compared with the prior art, the present invention has the following beneficial effects: In terms of data collection, the surveillance camera group adopts bionic vision principles and adaptive visual attention models, which can dynamically adjust the shooting angle and resolution according to the importance and motion status of the monitored target, thereby improving the pertinence and efficiency of data collection, ensuring that key targets are monitored with high resolution, while taking into account the surrounding environment and avoiding unnecessary data redundancy.

[0020] The wireless communication network innovatively introduces a hybrid communication mode that combines quantum encryption communication technology with traditional wireless communication technology. It intelligently selects the optimal communication method through a comprehensive evaluation formula, which not only ensures the efficiency of data transmission, but also greatly improves the security of data transmission, effectively preventing data from being stolen and tampered with.

[0021] The edge computing unit is equipped with an edge computing platform based on neuromorphic computing chips, which uses a pulse neural network algorithm for fast and low-power data processing. It can extract key features locally, reduce data transmission volume, reduce the processing burden of cloud data centers, and improve the response speed of the system.

[0022] The cloud data center adopts an architecture that combines distributed blockchain and federated learning, combined with a deep reinforcement transfer learning algorithm to ensure the immutability and traceability of the data. At the same time, it collaboratively trains models without sharing the original data, explores the potential information behind the data, and provides more accurate and comprehensive support for monitoring decisions.

[0023] The user interaction terminal has holographic projection interaction and emotion perception functions. It provides personalized services based on the user's emotional state and needs through a personalized recommendation model, enabling users to obtain monitoring information and make decisions more intuitively and conveniently, thereby improving the user experience.

[0024] The coordinated work of functional modules such as the intelligent behavior analysis module, energy management module, security encryption module, adaptive image enhancement module, multi-sensor fusion module and system self-diagnosis and optimization module further enhances the intelligence, energy saving, safety and reliability of the system. The intelligent behavior analysis module can accurately predict abnormal target behavior, the energy management module effectively reduces system energy consumption, the security encryption module ensures data security, the adaptive image enhancement module improves image quality, the multi-sensor fusion module provides more comprehensive environmental information, and the system self-diagnosis and optimization module promptly discovers and solves system faults to ensure the stable operation of the system.

[0025] To sum up, this patented technical solution has significantly improved the performance, security and user experience of the camera monitoring system through various innovations and optimizations, and has broad application prospects and market value. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 A schematic block diagram of a camera monitoring system and a communication protocol optimization system based on wireless communication proposed by the present invention; Figure 2 A schematic block diagram of a camera monitoring system based on wireless communication and a communication protocol optimization method proposed in the present invention; Figure 3 A comparison chart of data transmission performance under different communication modes of the camera monitoring system based on wireless communication and the communication protocol optimization method proposed in the present invention; Figure 4 This is a comparison chart of the data volume before and after processing by the edge computing unit of the camera monitoring system based on wireless communication and the communication protocol optimization method proposed in the present invention; Figure 5 This is a graph showing the change over time of the abnormal behavior warning accuracy of the intelligent behavior analysis module of the camera monitoring system based on wireless communication and the communication protocol optimization method proposed in the present invention; Figure 6 This is a graph showing the proportions of different power supply modes for the energy management module of the camera monitoring system based on wireless communication and the communication protocol optimization method proposed in the present invention. DETAILED DESCRIPTION

[0027] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0028] In the description of the present invention, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", "clockwise", "counterclockwise" and the like indicate orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the referred device or element must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as limiting the present invention.

[0029] In addition, the terms "first" and "second" are used for descriptive purposes only and cannot be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, the meaning of "multiple" is two or more, unless otherwise clearly and specifically defined. In addition, the terms "installed", "connected" and "connected" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, and it can be the internal connection of two elements. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances. The present invention will be further described in detail below in conjunction with the accompanying drawings.

[0030] Reference Figure 1-6 :A camera monitoring system based on wireless communication, including the following modules: Surveillance camera group: A camera array that uses the bionic vision principle to simulate the visual perception mechanism of the human eye and uses an adaptive visual attention model Dynamically adjust the shooting angle and resolution, including is the i-th visual feature, is the weight of the corresponding feature, and n is the visual collection sample; the system will comprehensively consider multiple visual feature samples. For example, when a fast-moving object appears in the monitoring area, the weight of the visual feature related to the movement will automatically increase, and the camera will quickly adjust the viewing angle to the object and increase the shooting resolution of the area to capture the details of the object more clearly; while for relatively static and less important background areas, the resolution will be appropriately reduced, thereby effectively saving storage space and transmission bandwidth while ensuring the quality of key information collection Wireless communication network: A hybrid communication mode combining quantum encryption communication technology and traditional wireless communication technology is introduced; during data transmission, according to the real-time signal strength S, bandwidth requirement B, transmission distance D and security level requirement L, a comprehensive evaluation formula is used to evaluate the Select the best communication method, where is the weight coefficient; when the signal strength in the monitoring area is weak, but the security requirement for data transmission is extremely high, the system will appropriately increase the weight coefficient of the security level requirement L , quantum encryption communication technology is preferred, and other technical means are combined to ensure the stability of transmission; in scenarios with good signals and high bandwidth requirements, the weight coefficient of bandwidth requirement B will be increased. , choose traditional wireless communication technology with faster transmission speed, and use quantum encryption technology to ensure data security Edge computing unit: Equipped with an edge computing platform based on neuromorphic computing chips, it processes data by imitating the working mode of human brain neurons; by building a feature extraction model Complete the feature extraction task, where is the jth pulse feature, is the weight of the corresponding feature; when the camera collects image data, the edge computing platform will quickly identify various features in the image, such as the outline and texture of the object, in the same way that the human brain processes visual information. Compared with traditional data processing methods, this method based on neuromorphic computing can process large amounts of monitoring data more quickly, reducing the pressure of data transmission to the cloud. At the same time, it can also analyze and process data locally in real time, detect abnormal situations in a timely manner and respond, greatly improving the real-time and intelligent level of the monitoring system. Cloud data center: Adopts a distributed blockchain and federated learning architecture, uses deep reinforcement transfer learning algorithm to analyze data, and uses reward function to Optimize model parameters, where is the reward value of the tth time step; in terms of architecture, the combination of distributed blockchain and federated learning greatly improves data security and collaboration. Distributed blockchain makes it difficult for data to be tampered with and forged, ensuring the credibility of monitoring data as evidence, which is of great significance in scenarios such as judicial evidence collection and security audits. Federated learning promotes multi-party data integration and breaks down data silos. Data centers in different regions or institutions can work together to improve model performance and tap into greater data value.

[0031] Deep reinforcement transfer learning algorithms can accurately mine potential information from monitoring data. They can detect abnormal behavior trends in advance to help with security precautions; they can analyze the flow of people and logistics and optimize resource allocation. For example, in the security of large-scale events, they can predict risk points where crowds gather; in logistics parks, they can optimize cargo transportation routes. All of these provide strong support for monitoring decisions and improve the system's intelligence level and practical application value.

[0032] User interaction terminal: Through holographic projection technology, emotion recognition sensors and natural language processing technology, it can perceive the user's emotional state and language intention, and use personalized recommendation models Automatically adjust the interactive interface and provide personalized services, including is the user's preference feature, is the weight of the corresponding feature. Holographic projection interaction enables users to view the monitoring screen in an all-round and immersive way, significantly improving the efficiency of information acquisition. In complex scenes such as traffic hub monitoring, users can quickly grasp the overall layout and details, making it easier to find abnormalities in a timely manner. Compared with traditional two-dimensional display, it can reduce visual fatigue and information omissions, and improve user experience.

[0033] Emotional perception and personalized recommendation functions make the interaction more intelligent and considerate. It can automatically adjust the interface and services according to the user's emotions and needs, and quickly provide key information in an emergency to assist users in making efficient decisions. Long-term use can continuously learn user preferences, provide more accurate services, enhance user dependence and satisfaction with the system, and enable the monitoring system to better serve the user's actual work and life.

[0034] The present invention also includes the following modules: Intelligent behavior analysis module: Based on the model of the fusion of spatiotemporal graph convolutional network ST-GCN and generative adversarial network GAN, it analyzes the target behavior trajectory and action mode in the monitoring area; Represent the spatiotemporal relationship of the target and use ST-GCN to extract spatiotemporal features ; The generator G in GAN attempts to generate behavior patterns, and the discriminator D distinguishes between real and generated behavior patterns. Through the behavior feature vector V and time series T, the formula Calculate the probability P of target behavior abnormality and issue an early warning when P exceeds the preset threshold. is the weight coefficient, and b is the bias term. This module is based on a model that combines the spatiotemporal graph convolutional network with the generative adversarial network, and can deeply analyze the behavior of targets in the monitoring area. By constructing a spatiotemporal graph to accurately present the spatiotemporal relationship of the target, combined with advanced algorithms to extract features, it effectively improves the ability to understand and identify complex behavior patterns. For example, in crowded places such as large shopping malls and airports, suspicious behaviors such as abnormal wandering and sudden running can be quickly identified; in industrial production monitoring, workers' irregular operations can be discovered in a timely manner.

[0035] When the calculated probability of abnormal target behavior exceeds the preset threshold, an early warning is issued, which provides strong support for security prevention work. It can help security personnel respond in time to prevent crimes and safety accidents; in corporate production, it can reduce production failures and safety hazards caused by improper operation. The early warning mechanism changes passive response to active prevention, greatly enhancing the security and practicality of the monitoring system, and playing an important role in ensuring the safety of personnel and property. Energy management module: A composite power supply method combining solar panels, rechargeable batteries and wireless energy collection technology; by monitoring the remaining battery power , Solar panel power generation , Wireless Energy Harvesting Power and system power consumption , using the formula Calculate the battery life T; when T is lower than the set value, the system uses the energy allocation model to allocate energy based on device priority and task importance. Automatically adjust the system working mode, including is the energy consumption requirement of the ith device, The priority weight of the corresponding device. The composite power supply method combining solar panels, rechargeable batteries and wireless energy collection technology broadens the energy acquisition channels and reduces the dependence on traditional electricity, which is both energy-saving and environmentally friendly and reduces operating costs. It is especially suitable for monitoring scenarios in remote areas or those that are difficult to access the power grid to ensure continuous and stable operation of the system.

[0036] By accurately monitoring parameters such as power and electricity and calculating battery life, when battery life is insufficient, the system automatically adjusts the working mode based on device priority and task importance to prioritize the operation of key equipment. For example, in outdoor ecological monitoring, priority is given to ensuring the power supply of cameras and key communication equipment to continuously record important ecological data; in traffic monitoring, priority is given to maintaining the normal operation of intersection cameras to ensure the effectiveness of traffic management. This intelligent energy allocation strategy improves energy efficiency, extends the use time of equipment, and enhances the reliability and stability of the monitoring system. Security encryption module: adopts an encryption method that combines quantum key distribution QKD with homomorphic encryption; QKD is used to generate and distribute quantum keys, and homomorphic encryption allows data processing in the ciphertext state. At the data sending end, the data is first encrypted once with the quantum key generated by QKD, and then the encrypted data is encrypted again using the homomorphic encryption algorithm; the receiving end performs the corresponding reverse decryption operation; the quantum key is further obfuscated by introducing a chaotic sequence encryption algorithm to enhance the security of the key. The encryption formula is: ,in is the original quantum key, C is the chaotic sequence, and f is the chaotic encryption function. A multi-level encryption method combining quantum key distribution and homomorphic encryption is used to build a solid defense line for secure data transmission and storage. The absolutely secure key generated by quantum key distribution fundamentally guarantees the initial security of the communication link and prevents the key from being stolen and cracked. Homomorphic encryption allows data to be processed in the ciphertext state, avoiding security risks in the decryption process. It can effectively protect sensitive information in some scenarios with extremely high requirements for data privacy, such as financial monitoring and medical monitoring data transmission.

[0037] The introduction of chaotic sequence encryption algorithm further confuses quantum keys, enhances the security of keys, and makes data almost impossible to crack during transmission and storage. This not only protects the privacy and integrity of monitoring data, but also improves the credibility of the entire monitoring system, allowing users to use the system to transmit and store important data with confidence, especially in scenarios involving legal evidence collection, commercial secret monitoring, etc., providing reliable protection for data security. Adaptive image enhancement module: The module combines the Retinex algorithm with the generative adversarial network GAN to enhance the image; according to the ambient light intensity L, image clarity index C, contrast index K and image noise level N, the formula ),in This is the image processed by the Retinex algorithm. Optimized images generated by GAN, is the weight coefficient, and Dynamically adjust camera image acquisition parameters and evaluate image quality through image quality indicators ,in is the weight coefficient. This module combines the multi-scale Retinex algorithm with the generative adversarial network to effectively improve the quality of surveillance images. The multi-scale Retinex algorithm enhances image brightness and color, and GAN optimizes detail textures to make images clearer and more realistic. In low-light or complex lighting environments, such as nighttime streets and dimly lit areas indoors, it can enhance image visibility and facilitate identification of target features.

[0038] By dynamically adjusting camera parameters and evaluating image quality in real time through formulas, image effects can be optimized in a timely manner according to environmental changes. This helps security personnel to clearly view surveillance images and detect potential threats; it also helps the intelligent analysis module to accurately identify target behaviors, improve the effectiveness and practicality of the surveillance system, and provide more reliable image basis for security prevention and decision-making. Multi-sensor fusion module: Integrates temperature, humidity, smoke, motion, and gas concentration sensors, and introduces a sensor data fusion algorithm based on a Bayesian network; the algorithm builds a Bayesian network model The data from different sensors are analyzed and fused, where G is a directed acyclic graph, indicating the dependency relationship between sensors. is the conditional probability distribution), according to the sensor data , using the Bayesian formula Calculate the probability of different events, where H is the hypothetical event, is the prior probability of the hypothesized event, is the probability of sensor data occurring when the hypothetical event occurs, The probability of sensor data appearing. By integrating multiple sensors and introducing the Bayesian network algorithm, environmental information can be fully perceived. Different sensors collect data from multiple dimensions such as temperature, humidity, and smoke, and the Bayesian network algorithm integrates and analyzes data to improve the monitoring ability of complex environments. In large commercial buildings or industrial sites, abnormal events such as fires and intrusions can be discovered in a timely manner.

[0039] By calculating the probability of different events, early warnings are generated, which is more accurate and comprehensive than single sensors. Early warnings allow relevant personnel to take timely measures to reduce accident losses; at the same time, they provide rich data support for intelligent decision-making, improve the security and intelligence level of the monitoring system, and ensure the safety of personnel and property. System self-diagnosis and optimization module: Adopt PHM technology based on fault prediction and health management; regularly test the system hardware performance, software operation status and communication quality, build a deep convolutional autoencoder DCAE model to extract features and detect anomalies in the system operation data, and compare the feature vectors under normal conditions. and the feature vector at the current state , using the similarity measurement formula Calculate the similarity S; when S is lower than the set threshold, determine that the system may have a fault; based on the detection result R and the historical optimization record H, use the reinforcement learning algorithm to generate the optimization strategy O, and use the reward function Automatically adjust system parameters to improve overall performance, including is the reinforcement learning reward for the i-th time step. The fault prediction and health management technology based on deep learning can be used to regularly detect the system status. The deep convolutional autoencoder extracts features and detects anomalies, timely discovers hardware, software and communication problems, and avoids monitoring interruptions caused by system failures.

[0040] By using similarity measurement to judge faults, using reinforcement learning algorithms to generate optimization strategies, and automatically adjusting system parameters, this can improve system performance, extend service life, and reduce manual maintenance costs. In long-term operation, the system is continuously optimized to ensure that the monitoring system is stable and reliable, providing a solid guarantee for security monitoring work, especially for scenarios with high requirements for system stability. The present invention also discloses a communication protocol optimization method for a camera monitoring system based on wireless communication, comprising the following steps: Data collection steps: The surveillance camera group uses the bionic vision principle and adaptive visual attention model to dynamically adjust the shooting angle and resolution according to the importance and motion state of the monitoring target, and collect audio and video data in the monitoring area; this step is based on the bionic vision principle and adaptive model, and can dynamically adjust the shooting settings according to the monitoring target. This makes the collected audio and video data highly targeted and rich in information, effectively avoiding the collection of useless data and reducing data redundancy. In complex and changing monitoring scenarios, key target information can be clearly captured, providing high-quality original materials for subsequent data processing and analysis, and improving monitoring efficiency and accuracy.

[0041] Communication selection steps: The wireless communication network uses a comprehensive evaluation formula to select a communication method that combines quantum encryption communication technology with traditional wireless communication technology to transmit data collected by the surveillance camera group based on the real-time signal strength S, bandwidth requirement B, transmission distance D, and security level requirement L; the wireless communication network intelligently selects the optimal communication method by comprehensively evaluating multiple factors. This ensures efficient and secure data transmission under different signal, bandwidth, distance, and security requirements. It avoids data transmission delays, packet loss, or security vulnerabilities caused by improper communication methods, ensures that monitoring data reaches the processing end in a timely and complete manner, and lays the foundation for the stable operation of the monitoring system.

[0042] Edge processing steps: The edge computing unit uses a feature extraction model based on neuromorphic computing chips and pulse neural network (SNN) algorithms to perform preliminary processing on the collected data and extract key features; the edge computing unit uses advanced chips and algorithms to perform preliminary processing on the collected data. Rapidly extracting key features and reducing the amount of transmission reduces data transmission pressure and latency. In monitoring scenarios with large amounts of data, effective information can be transmitted to the cloud faster, while reducing energy consumption and improving the overall operating efficiency of the system, allowing the cloud to focus more on in-depth analysis and improve the response speed and performance of the monitoring system.

[0043] Cloud analysis steps: The cloud data center adopts a distributed blockchain and federated learning architecture, and uses the reinforcement transfer learning algorithm and reward function to analyze the received data to provide support for monitoring decisions; the cloud data center uses a unique architecture and advanced algorithms to deeply mine the potential information of the data. Distributed blockchain and federated learning ensure data security and multi-party collaboration, and the deep reinforcement transfer learning algorithm accurately analyzes the data. It can discover abnormal behaviors, trends and other information in the monitoring scene, provide strong support for users to make scientific monitoring decisions, and enhance the intelligence and practicality of the monitoring system.

[0044] Result feedback step: The user interaction terminal uses holographic projection interaction and emotion perception functions to perceive the user's emotional state and language intention according to the personalized recommendation model, and feeds back the analysis results to the user through the cloud data center. The user interaction terminal uses holographic projection and emotion perception to provide personalized feedback on analysis results. It can adjust the feedback method according to the user's state and intention, allowing users to obtain the required monitoring information more conveniently and improve the interactive experience. It helps users quickly understand complex monitoring data and make accurate decisions, especially in emergency situations, which can improve response efficiency and give full play to the value of the monitoring system.

[0045] Security encryption step: The security encryption module uses an encryption method that combines quantum key distribution QKD with homomorphic encryption, and uses a chaotic sequence encryption algorithm to confuse quantum keys to ensure the security of data transmission and storage; this step uses a multi-level encryption method and a chaotic sequence algorithm to confuse quantum keys, greatly enhancing the security of data during transmission and storage. It can effectively resist various high-level hacker attacks and data theft, ensure the privacy and integrity of monitoring data, and build a strong defense line for data security in monitoring scenarios involving sensitive information.

[0046] Image enhancement steps: The adaptive image enhancement module combines the Retinex algorithm with the generative adversarial network GAN. According to the ambient light intensity L, image clarity index C, contrast index K, and image noise level N, the camera image acquisition parameters are dynamically adjusted through formulas and image quality evaluation indicators to improve image quality. By combining the multi-scale Retinex algorithm with the generative adversarial network, the acquisition parameters are dynamically adjusted according to the environment and image indicators, which can significantly improve image quality. Even in complex lighting conditions and poor image quality, the picture clarity and contrast can be enhanced, making it easier for monitoring personnel to identify targets more clearly. At the same time, it provides high-quality image data for intelligent analysis and improves the efficiency of the monitoring system.

[0047] Behavior analysis steps: The intelligent behavior analysis module is based on a model that combines the spatiotemporal graph convolutional network ST-GCN with the generative adversarial network GAN. It uses a formula to analyze the probability of abnormal target behavior and issues an early warning when the probability exceeds the threshold. Based on the fusion model and specific formula, the probability of abnormal target behavior can be analyzed to detect abnormal behavior in the monitoring area in a timely manner. In the monitoring of public places or key facilities, potential security threats can be warned in advance, helping security personnel to respond quickly, preventing dangerous incidents, and enhancing the safety and stability of the monitoring area.

[0048] Energy management steps: The energy management module uses a composite power supply method that combines solar panels, rechargeable batteries and wireless energy collection technology to calculate the battery life T according to the formula; when T is lower than the set value, the system working mode is automatically adjusted through the energy allocation model to reduce the overall energy consumption; the composite power supply method and the intelligent energy management of relevant formulas can effectively extend the battery life. When the battery life is insufficient, the system working mode is automatically adjusted to give priority to the operation of key equipment and reduce the overall energy consumption. It is suitable for monitoring scenarios where power supply is inconvenient in the wild and remote areas, ensuring the long-term stable operation of the monitoring system and reducing maintenance costs.

[0049] Data fusion steps: sensor fusion module integration, using the sensor data fusion algorithm based on Bayesian network and Bayesian formula to correlate and fuse the data of different sensors to generate environmental information and monitoring warnings; using multi-sensor fusion and Bayesian network algorithm to correlate and fuse different sensor data, it can generate more comprehensive and accurate environmental information and monitoring warnings. In complex environment monitoring, multi-dimensional data can be integrated to timely discover potential dangers such as fire and intrusion, improve the accuracy and reliability of the monitoring system, and provide strong support for safety decision-making.

[0050] System optimization steps: The system self-diagnosis and optimization module uses PHM technology based on system fault prediction and health management, detects system faults through similarity measurement formulas, and generates and executes optimization strategies using learning algorithms and reward functions. Detecting system faults based on deep learning technology and generating optimization strategies through reinforcement learning can timely discover and solve potential system problems. Continuously optimize system performance, reduce the probability of failure, and ensure stable and efficient operation of the monitoring system, especially in long-term uninterrupted monitoring tasks, which can effectively improve the reliability and service life of the system.

[0051] The above are only preferred specific implementation modes of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical solutions and inventive concepts of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A camera monitoring system based on wireless communication, characterized in that: Includes the following modules: Surveillance camera group: A camera array that uses the bionic vision principle to simulate the visual perception mechanism of the human eye and uses an adaptive visual attention model Dynamically adjust the shooting angle and resolution, including is the i-th visual feature, is the weight of the corresponding feature, n is the visual collection sample; Wireless communication network: A hybrid communication mode combining quantum encryption communication technology and traditional wireless communication technology is introduced; during data transmission, according to the real-time signal strength S, bandwidth requirement B, transmission distance D and security level requirement L, a comprehensive evaluation formula is used to evaluate the Select the best communication method, where is the weight coefficient; Edge computing unit: Equipped with an edge computing platform based on neuromorphic computing chips, it processes data by imitating the working mode of human brain neurons; by building a feature extraction model Complete the feature extraction task, where is the jth pulse feature, is the weight of the corresponding feature; Cloud data center: Adopts a distributed blockchain and federated learning architecture, uses a reinforcement transfer learning algorithm to analyze data, and uses a reward function to Optimize model parameters, where is the reward value at the tth time step; User interaction terminal: Through holographic projection technology, emotion recognition sensors and natural language processing technology, it can perceive the user's emotional state and language intention, and use personalized recommendation models Automatically adjust the interactive interface and provide personalized services, including is the user's preference feature, is the weight of the corresponding feature.

2. The camera monitoring system based on wireless communication according to claim 1, characterized in that: Also includes: Intelligent behavior analysis module: Based on the model of the fusion of spatiotemporal graph convolutional network ST-GCN and generative adversarial network GAN, it analyzes the target behavior trajectory and action mode in the monitoring area; Represent the spatiotemporal relationship of the target and use ST-GCN to extract spatiotemporal features ; The generator G in GAN attempts to generate behavior patterns, and the discriminator D distinguishes between real and generated behavior patterns. Through the behavior feature vector V and time series T, the formula Calculate the probability P of target behavior abnormality and issue an early warning when P exceeds the preset threshold. is the weight coefficient and b is the bias term.

3. The camera monitoring system based on wireless communication according to claim 1, characterized in that: Also includes: Energy management module: A composite power supply method that combines solar panels, rechargeable batteries and wireless energy collection technology; By monitoring the remaining battery power , Solar panel power generation , Wireless Energy Harvesting Power and system power consumption , using the formula Calculate the battery life T; when T is lower than the set value, the system uses the energy allocation model to allocate energy based on device priority and task importance. Automatically adjust the system working mode, including is the energy consumption requirement of the ith device, is the priority weight of the corresponding device.

4. The camera monitoring system based on wireless communication according to claim 1, characterized in that: Also includes: Security encryption module: an encryption method combining quantum key distribution (QKD) and homomorphic encryption; QKD is used to generate and distribute quantum keys, while homomorphic encryption allows data processing in a ciphertext state. At the data sending end, the data is first encrypted once with the quantum key generated by QKD, and then the encrypted data is re-encrypted using the homomorphic encryption algorithm; the receiving end performs the corresponding reverse decryption operation; the quantum key is obfuscated by introducing a chaotic sequence encryption algorithm to enhance the security of the key. The encryption formula is: ,in is the original quantum key, C is the chaotic sequence, and f is the chaotic encryption function.

5. The camera monitoring system based on wireless communication according to claim 1, characterized in that: Also includes: Adaptive image enhancement module: The module combines the Retinex algorithm with the generative adversarial network GAN for image enhancement; According to the ambient light intensity L, image clarity index C, contrast index K and image noise level N, the formula is used ),in This is the image processed by the Retinex algorithm. Optimized images generated by GAN, is the weight coefficient, and Dynamically adjust camera image acquisition parameters and evaluate image quality through image quality indicators ,in is the weight coefficient.

6. The camera monitoring system based on wireless communication according to claim 1, characterized in that: Also includes: Multi-sensor fusion module: Integrates temperature, humidity, smoke, motion, and gas concentration sensors, and introduces a sensor data fusion algorithm based on a Bayesian network; the algorithm builds a Bayesian network model The data from different sensors are analyzed and fused, where G is a directed acyclic graph, indicating the dependency relationship between sensors. is the conditional probability distribution), according to the sensor data , using the Bayesian formula Calculate the probability of different events, where H is the hypothetical event, is the prior probability of the hypothesized event, is the probability of sensor data occurring when the hypothetical event occurs, is the probability of sensor data occurrence.

7. The camera monitoring system based on wireless communication according to claim 1, characterized in that: Also includes: System self-diagnosis and optimization module: adopts fault prediction and health management PHM technology based on system learning; regularly checks the system hardware performance, software operation status and communication quality, builds a deep convolutional autoencoder DCAE model to extract features and detect anomalies in the system operation data, and compares the feature vectors in the normal state. and the feature vector at the current state , using the similarity measurement formula Calculate the similarity S; When S is lower than the set threshold, it is determined that the system may have a fault; based on the detection result R and the historical optimization record H, the reinforcement learning algorithm is used to generate the optimization strategy O, and the reward function Automatically adjust system parameters to improve overall performance, including is the reinforcement learning reward at the i-th time step.

8. A communication protocol optimization method for a camera monitoring system based on wireless communication using any one of claims 1 to 7, characterized in that: The following steps are involved: Data collection steps: The surveillance camera group uses bionic vision principles and adaptive visual attention models to dynamically adjust the shooting angle and resolution according to the importance and motion status of the monitored target, and collect audio and video data in the monitored area; Communication selection step: The wireless communication network uses a comprehensive evaluation formula to select a communication method that combines quantum encryption communication technology with traditional wireless communication technology based on real-time signal strength S, bandwidth requirement B, transmission distance D, and security level requirement L to transmit data collected by the surveillance camera group; Edge processing steps: The edge computing unit uses a feature extraction model based on a neuromorphic computing chip and a spiking neural network (SNN) algorithm to preliminarily process the collected data and extract key features; Cloud analysis step: The cloud data center uses a distributed blockchain and federated learning architecture to analyze the received data using a reinforcement transfer learning algorithm and reward function to provide support for monitoring decisions; Result feedback step: The user interaction terminal uses holographic projection interaction and emotion perception functions to perceive the user's emotional state and language intention based on the personalized recommendation model, and feeds back the analysis results to the user through the cloud data center.

9. The communication protocol optimization method according to claim 8, characterized in that: Also includes: Security encryption steps: The security encryption module uses an encryption method that combines quantum key distribution QKD with homomorphic encryption, and uses a chaotic sequence encryption algorithm to confuse the quantum key to ensure the security of data transmission and storage; Image enhancement step: The adaptive image enhancement module combines the Retinex algorithm with the generative adversarial network GAN to dynamically adjust the camera image acquisition parameters through formulas and image quality evaluation indicators according to the ambient light intensity L, image clarity index C, contrast index K, and image noise level N to improve image quality; Behavior analysis steps: The intelligent behavior analysis module is based on a model that combines the spatiotemporal graph convolutional network ST-GCN with the generative adversarial network GAN. It uses a formula to analyze the probability of abnormal target behavior and issues an early warning when the probability exceeds the threshold.

10. The communication protocol optimization method according to claim 8, characterized in that: Also includes: Energy management steps: The energy management module uses a composite power supply method that combines solar panels, rechargeable batteries and wireless energy collection technology to calculate the battery life T according to the formula; when T is lower than the set value, the system working mode is automatically adjusted through the energy allocation model to reduce overall energy consumption; Data fusion step: sensor fusion module integration, using the sensor data fusion algorithm based on Bayesian network and Bayesian formula to perform correlation analysis and fusion on the data of different sensors to generate environmental information and monitoring warning; System optimization steps: The system self-diagnosis and optimization module adopts PHM technology based on system fault prediction and health management, detects system faults through similarity measurement formula, and uses learning algorithms and reward functions to generate and execute optimization strategies.

Images