Signal transmission method and system for infrared thermal imaging core

Abstract

The present invention relates to the field of infrared image processing, and discloses a signal transmission method and system for an infrared thermal imaging core, the method comprising obtaining initial infrared thermal imaging data, network environment parameters and packet loss information; adaptively and dynamically adjusting the transmission strategy according to the initial infrared thermal imaging data and the network environment parameters to obtain an optimized transmission strategy; analyzing the packet loss mechanism according to the packet loss information to obtain a packet loss mapping relationship model; dynamically balancing image quality and transmission real-time according to the network environment parameters, the optimized transmission strategy and the packet loss mapping relationship model to obtain a balance constraint condition; estimating future bandwidth requirements according to the balance constraint condition to obtain a network resource pre-allocation scheme; establishing a closed-loop optimization model according to the network resource pre-allocation scheme and performing model optimization iteration to obtain a stable state transmission scheme. The present method can realize adaptive and dynamically adjusted infrared thermal imaging core data transmission.

Description

Technical Field

[0001] The present invention relates to the field of infrared image processing, and in particular to a signal transmission method and system for an infrared thermal imaging core. Background Art

[0002] At present, in the field of signal transmission of infrared thermal imaging cores, with the rapid development of infrared thermal imaging technology, the resolution and refresh rate are constantly improving, resulting in a sharp increase in the amount of raw data. For example, the data volume of an infrared thermal imaging video stream with a resolution of 4K and 30 frames per second can reach hundreds of megabytes or even gigabytes per second when uncompressed. However, the network bandwidth resources in actual application scenarios are limited, and the network environment is complex and changeable, with problems such as bandwidth fluctuations, delay jitter, and packet loss. In addition, the requirements for image quality and real-time transmission in different application scenarios vary significantly. For example, image quality is crucial in industrial inspection scenarios, while security monitoring scenarios pay more attention to real-time transmission. How to dynamically balance image quality and real-time transmission under limited network resources, while achieving network adaptive adjustment, has become the core challenge of signal transmission for infrared thermal imaging cores.

[0003] In one prior art, the signal transmission of an infrared thermal imaging movement usually adopts a fixed compression ratio and a static transmission strategy. Specifically, the prior art compresses the original infrared thermal imaging data through a preset compression algorithm, and then transmits it with a fixed data packet size and sending interval. The network status monitoring module is only used to feedback network bandwidth and delay information in real time, but lacks in-depth analysis and dynamic adjustment mechanism of the network status. For example, when the network is congested or the signal is attenuated, the prior art cannot adjust the compression ratio or transmission strategy in real time, resulting in an increase in packet loss rate or a significant increase in transmission delay. In addition, a single image quality evaluation indicator is used to judge the transmission effect, while ignoring the differentiated requirements of image quality and real-time performance in different application scenarios. Therefore, it is difficult to achieve adaptive and dynamically adjusted infrared thermal imaging movement data transmission in the prior art. Summary of the invention

[0004] The present invention provides a signal transmission method and system for an infrared thermal imaging core, so as to realize adaptive and dynamically adjusted infrared thermal imaging core data transmission.

[0005] In a first aspect, in order to solve the above technical problems, the present invention provides a signal transmission method of an infrared thermal imaging core, comprising:

[0006] Obtain initial infrared thermal imaging data, network environment parameters and packet loss information;

[0007] According to the initial infrared thermal imaging data and the network environment parameters, a fuzzy clustering algorithm is used to classify the network status, and the transmission strategy is adaptively and dynamically adjusted to obtain an optimized transmission strategy;

[0008] According to the packet loss information, a data mining algorithm is used to analyze the packet loss mechanism and obtain a packet loss mapping relationship model;

[0009] According to the network environment parameters, the optimized transmission strategy and the packet loss mapping relationship model, combined with the application scenario requirements, dynamically balance the image quality and the real-time transmission to obtain the balance constraint condition;

[0010] According to the balance constraint condition, the network environment parameter and the packet loss mapping relationship model, a time series prediction algorithm is used to estimate future bandwidth demand to obtain a network resource pre-allocation plan;

[0011] According to the network resource pre-allocation scheme and the network environment parameters, a closed-loop optimization model is established in combination with transmission efficiency and cost constraints, and model optimization iterations are performed to obtain a steady-state transmission scheme.

[0012] In an optional implementation, the obtaining of initial infrared thermal imaging data, network environment parameters and packet loss information includes:

[0013] Obtain original infrared thermal imaging data, real-time feedback of network environment parameters and packet loss information during network transmission;

[0014] According to the original infrared thermal imaging data, a data compression algorithm is used in combination with a preset compression ratio threshold to perform spatial redundancy compression processing to obtain original compressed data;

[0015] Calculating based on the original compressed data and the network environment parameters to obtain a maximum transmission data volume threshold corresponding to the current network bandwidth;

[0016] The data volume of the original compressed data is compared with the maximum transmission data volume threshold. When the data volume of the original compressed data is greater than the maximum transmission data volume threshold, the compression ratio threshold is dynamically increased and the compression process is re-executed to obtain the initial infrared thermal imaging data; when the data volume of the original compressed data is less than the maximum transmission data volume threshold, the original compressed data is determined to be the initial infrared thermal imaging data.

[0017] In an optional implementation, the method of classifying the network status using a fuzzy clustering algorithm based on the initial infrared thermal imaging data and the network environment parameters, and adaptively and dynamically adjusting the transmission strategy to obtain an optimized transmission strategy includes:

[0018] According to the network environment parameters, a fuzzy clustering algorithm is used to perform a three-dimensional network status feature space analysis including signal strength, bandwidth fluctuation rate and round-trip delay and divide the network status to obtain a real-time network status stability level;

[0019] According to the initial infrared thermal imaging data, key frame distribution characteristics and temperature sensitive area analysis are performed to obtain a transmission priority weight matrix;

[0020] The historical network state stability level and the historical transmission priority weight matrix are used as input, and the historical optimized transmission strategy is used as output, a transmission strategy adaptive dynamic adjustment model is constructed and trained, and the training is determined to be completed when the number of training times is greater than or equal to the preset number of training times, and a transmission strategy adaptive dynamic adjustment model that has completed training is obtained;

[0021] The network state stability level and the transmission priority weight matrix are input into the trained transmission strategy adaptive dynamic adjustment model to obtain an optimized transmission strategy.

[0022] In an optional implementation, the packet loss mechanism analysis is performed using a data mining algorithm according to the packet loss information to obtain a packet loss mapping relationship model, including:

[0023] According to the packet loss information, extract the data compression type of each packet loss event and the key frame identifier of the corresponding image frame to obtain a data packet attribute feature matrix;

[0024] According to the packet loss information, analyzing the timestamp information of the packet loss event, the network node location identifier and the lost data packet sequence number to obtain a spatiotemporal feature vector;

[0025] According to the packet loss information, a temperature matrix coordinate region analysis corresponding to the lost data packet is performed, and a maximum temperature deviation value of the region is calculated to obtain a quantitative index of image quality impact;

[0026] According to the data packet attribute feature matrix, the spatiotemporal feature vector and the image quality impact quantitative index, a data mining algorithm is used to analyze the packet loss mechanism to obtain a dynamically adjusted packet loss mapping relationship model.

[0027] In an optional implementation, the image quality and transmission real-time performance are dynamically balanced according to the network environment parameters, the optimized transmission strategy, and the packet loss mapping relationship model in combination with application scenario requirements to obtain a balance constraint condition, including:

[0028] According to the network environment parameters, the current network transmission bandwidth and the current network transmission delay are extracted, and the real-time performance of the transmission channel is combined for analysis to obtain a real-time image transmission bandwidth threshold;

[0029] According to the optimized transmission strategy, combined with a preset multi-scenario quality requirement feature library, quality requirement analysis is performed to obtain a peak signal-to-noise ratio threshold for each scenario and a maximum allowable transmission delay for each scenario;

[0030] According to the image transmission bandwidth threshold, the peak signal-to-noise ratio threshold of each scene and the maximum allowable transmission delay of each scene, a non-dominated sorting genetic algorithm is used to perform multivariate collaborative analysis to obtain a multi-objective optimization function including a compression rate influencing factor and a delay penalty factor;

[0031] According to the multi-objective optimization function and the packet loss mapping relationship model, an image quality assessment algorithm is used to dynamically balance image quality and transmission real-time performance to obtain a balance constraint condition;

[0032] Wherein, the multi-objective optimization function is as follows:

[0033]

[0034] in, represents the multi-objective optimization function, represents the image compression rate, represents the reference compression ratio, is the adjustment coefficient, is the transmission delay, is the delay penalty weight; Indicates ideal quality for each image area, Indicates The actual reception quality of each image area, represents the regional importance weight, The number of regions that the image is divided into, A coefficient for adjusting the weights of different target influences.

[0035] In an optional implementation, the estimating future bandwidth demand using a time series prediction algorithm according to the balance constraint condition, the network environment parameter and the packet loss mapping relationship model to obtain a network resource pre-allocation scheme includes:

[0036] According to the balance constraint condition and the network environment parameter, an image compression rate gradient change rule and a transmission delay compensation strategy are analyzed to obtain a bandwidth demand time series feature vector;

[0037] According to the network environment parameters, combined with a preset multi-scenario quality requirement feature library, a collaborative analysis of network state stability and scenario quality requirements is performed to obtain resource allocation weight coefficients for different transmission periods;

[0038] According to the packet loss mapping relationship model, network node location identification and packet loss event analysis are performed to obtain a channel quality assessment matrix;

[0039] According to the bandwidth demand timing characteristic vector, the resource allocation weight coefficient, the channel quality assessment matrix and the balance constraint condition, a time series prediction algorithm is used to pre-allocate network resources to obtain a network resource pre-allocation plan.

[0040] In an optional implementation, the establishing of a closed-loop optimization model based on the network resource pre-allocation scheme and the network environment parameters in combination with transmission efficiency and cost constraints and performing model optimization iterations to obtain a steady-state transmission scheme includes:

[0041] According to the network resource pre-allocation scheme, a multivariate collaborative analysis of bandwidth allocation gradient values, transmission priority weights, and channel quality assessment matrices is performed to obtain a multi-dimensional optimization input vector;

[0042] According to the network environment parameters, combined with a preset transmission efficiency cost function, calculation is performed to obtain a real-time transmission efficiency score and a real-time cost consumption index;

[0043] According to the multi-dimensional optimization input vector, a dynamic policy gradient algorithm is used to analyze and derive, and a closed-loop optimization strategy including a bandwidth adjustment coefficient, a compression ratio correction factor, and a transmission path selection parameter is obtained;

[0044] According to the closed-loop optimization strategy, network resources are reallocated and transmission strategies are iteratively updated, and the image quality score after transmission, the transmission delay after transmission, and the packet loss rate after transmission are monitored in real time to obtain strategy effect feedback data;

[0045] According to the feedback data of the strategy effect, a sliding window mechanism is used to perform spatiotemporal correlation analysis, identify the delayed effect of strategy adjustment on image quality and transmission efficiency, and obtain the strategy optimization deviation;

[0046] A closed-loop optimization model is constructed according to the strategy optimization deviation, the closed-loop optimization strategy, the real-time transmission efficiency score and the real-time cost consumption index, and the closed-loop optimization model is trained. When the decrease rate of the multi-objective loss function is lower than a preset threshold in a consecutive preset number of iterations, the current optimization model parameters are determined to be a steady-state transmission scheme.

[0047] In a second aspect, the present invention provides a signal transmission system for an infrared thermal imaging core, comprising:

[0048] Data acquisition module, used to obtain initial infrared thermal imaging data, network environment parameters and packet loss information;

[0049] A strategy optimization module, for classifying the network status using a fuzzy clustering algorithm according to the initial infrared thermal imaging data and the network environment parameters, and performing adaptive dynamic adjustment of the transmission strategy to obtain an optimized transmission strategy;

[0050] A packet loss analysis module, used to analyze the packet loss mechanism based on the packet loss information using a data mining algorithm to obtain a packet loss mapping relationship model;

[0051] A balance constraint module, used to dynamically balance image quality and transmission real-time performance according to the network environment parameters, the optimized transmission strategy and the packet loss mapping relationship model, combined with application scenario requirements, to obtain a balance constraint condition;

[0052] A resource allocation module, configured to estimate future bandwidth demand using a time series prediction algorithm according to the balance constraint condition, the network environment parameter and the packet loss mapping relationship model, and obtain a network resource pre-allocation plan;

[0053] The result output module is used to establish a closed-loop optimization model and perform model optimization iteration according to the network resource pre-allocation plan and the network environment parameters in combination with transmission efficiency and cost constraints to obtain a stable transmission plan.

[0054] In a third aspect, the present invention further provides an electronic device comprising a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein when the processor executes the computer program, it implements the signal transmission method of the infrared thermal imaging movement described in any one of the above.

[0055] In a fourth aspect, the present invention further provides a computer-readable storage medium, which includes a stored computer program, wherein when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute any one of the above-mentioned signal transmission methods for the infrared thermal imaging movement.

[0056] Compared with the prior art, the present invention has the following beneficial effects:

[0057] The invention discloses a signal transmission method and system for an infrared thermal imaging movement, comprising obtaining initial infrared thermal imaging data, network environment parameters and packet loss information; according to the initial infrared thermal imaging data and the network environment parameters, using a fuzzy clustering algorithm to classify the network state, and adaptively and dynamically adjust the transmission strategy to obtain an optimized transmission strategy; according to the packet loss information, using a data mining algorithm to analyze the packet loss mechanism and obtain a packet loss mapping relationship model; according to the network environment parameters, the optimized transmission strategy and the packet loss mapping relationship model, combined with application scenario requirements, dynamically balancing image quality and transmission real-time performance to obtain a balance constraint condition; according to the balance constraint condition, the network environment parameters and the packet loss mapping relationship model, using a time series prediction algorithm to estimate future bandwidth requirements to obtain a network resource pre-allocation plan; according to the network resource pre-allocation plan and the network environment parameters, combined with transmission efficiency and cost constraints, a closed-loop optimization model is established and model optimization iteration is performed to obtain a stable state transmission plan.

[0058] The present invention realizes efficient data transmission of infrared thermal imaging core in complex network environment by constructing adaptive dynamic adjustment signal transmission method. First, the initial infrared thermal imaging data, network environment parameters and packet loss information are obtained to provide basic data for subsequent transmission optimization. Then, in view of the dynamic change of network state, the fuzzy clustering algorithm is used to classify the network state, and the transmission strategy is adaptively and dynamically adjusted based on the classification result to obtain the optimized transmission strategy, thereby improving the adaptability of data transmission. Further, in order to deeply analyze the packet loss characteristics, a data mining algorithm is used to construct a packet loss mapping relationship model, and the packet loss mechanism is analyzed by extracting the data packet attribute characteristics, spatiotemporal characteristics and image quality impact quantitative indicators to support more refined transmission control. Next, combined with network environment parameters, optimized transmission strategy and packet loss mapping relationship model, comprehensive application scenario requirements, non-dominated sorting genetic algorithm is used to perform multi-objective optimization, dynamically balance image quality and transmission real-time, and form a balanced constraint condition. Based on the balanced constraint condition, a time series prediction algorithm is further used to predict future bandwidth requirements in combination with the bandwidth demand time series feature vector, resource allocation weight coefficient and channel quality assessment matrix, obtain a network resource pre-allocation scheme, and realize forward-looking resource allocation. Finally, based on the network resource pre-allocation scheme, combined with transmission efficiency and cost constraints, a closed-loop optimization model is constructed using a dynamic policy gradient algorithm, and optimization iterations are performed through the spatiotemporal correlation analysis of transmission feedback data. When the rate of decrease of the multi-objective loss function tends to be stable, a stable state transmission scheme is finally formed, thereby achieving long-term stable and efficient transmission. Therefore, the present invention realizes adaptive and dynamically adjusted infrared thermal imaging core data transmission, making it more stable and adaptable in complex network environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 It is a schematic flow chart of a signal transmission method of an infrared thermal imaging core provided by the first embodiment of the present invention;

[0060] Figure 2 It is a schematic diagram of the structure of the signal transmission system of the infrared thermal imaging core provided by the second embodiment of the present invention. DETAILED DESCRIPTION

[0061] 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.

[0062] Reference Figure 1 The first embodiment of the present invention provides a signal transmission method for an infrared thermal imaging core, comprising the following steps:

[0063] S11, obtaining initial infrared thermal imaging data, network environment parameters and packet loss information;

[0064] S12, classifying the network status using a fuzzy clustering algorithm according to the initial infrared thermal imaging data and the network environment parameters, and adaptively and dynamically adjusting the transmission strategy to obtain an optimized transmission strategy;

[0065] S13, using a data mining algorithm to analyze the packet loss mechanism according to the packet loss information to obtain a packet loss mapping relationship model;

[0066] S14, dynamically balancing image quality and transmission real-time performance according to the network environment parameters, the optimized transmission strategy, and the packet loss mapping relationship model in combination with application scenario requirements, to obtain a balance constraint condition;

[0067] S15, according to the balance constraint condition, the network environment parameter and the packet loss mapping relationship model, using a time series prediction algorithm to estimate future bandwidth demand and obtain a network resource pre-allocation plan;

[0068] S16, according to the network resource pre-allocation plan and the network environment parameters, combined with transmission efficiency and cost constraints, a closed-loop optimization model is established and model optimization iteration is performed to obtain a stable transmission plan.

[0069] In step S11, it is necessary to obtain initial infrared thermal imaging data, network environment parameters and packet loss information.

[0070] In one implementation, the obtaining of initial infrared thermal imaging data, network environment parameters, and packet loss information includes:

[0071] Original infrared thermal imaging data, real-time feedback of network environment parameters and packet loss information during network transmission are obtained; based on the original infrared thermal imaging data, a data compression algorithm is used in combination with a preset compression ratio threshold to perform spatial redundancy compression processing to obtain original compressed data; based on the original compressed data and the network environment parameters, a calculation is performed to obtain a maximum transmission data volume threshold corresponding to the current network bandwidth; the data volume of the original compressed data is compared with the maximum transmission data volume threshold, and when the data volume of the original compressed data is greater than the maximum transmission data volume threshold, the compression ratio threshold is dynamically increased and compression processing is re-executed to obtain initial infrared thermal imaging data; when the data volume of the original compressed data is less than the maximum transmission data volume threshold, the original compressed data is determined to be initial infrared thermal imaging data.

[0072] It should be noted that dynamically increasing the compression ratio threshold is an adaptive strategy that adjusts the compression ratio threshold in real time; it can adjust the compression strategy according to the actual data volume and network conditions to adapt to different transmission requirements, and ultimately make the data volume of the original compressed data less than the maximum transmission data volume threshold. The initial infrared thermal imaging data is adaptive transmission data that has undergone dynamic compression processing. It is obtained by obtaining the original thermal radiation signal, generating the original infrared image matrix through analog-to-digital conversion, using the frequency domain decomposition algorithm to process the image in blocks, and removing the high-frequency components according to the preset initial compression ratio threshold to complete the first round of data compression. The network environment parameters are collected in real time by the embedded network monitoring module, including three core indicators: current available bandwidth, data round-trip delay, and wireless signal strength. The current available bandwidth is dynamically obtained through the transport layer throughput test, and the data round-trip delay is calculated based on the round-trip time difference of the network protocol packet. The packet loss information includes the unique identifier of the lost data packet, the occurrence timestamp, and the transmission path node information, which comes from the transport layer data packet sequence analysis and is obtained by comparing the difference in the data packet sequence numbers between the sender and the receiver. The initial infrared thermal imaging data is used for subsequent network transmission optimization to ensure that better image quality is maintained under limited bandwidth. The network environment parameters are used to adjust the compression ratio, transmission priority and bandwidth allocation scheme in the transmission strategy optimization process to adapt to different network conditions and improve transmission stability. The packet loss information is analyzed by a data mining algorithm to form a packet loss mapping relationship model to help identify packet loss patterns and make dynamic adjustments in subsequent optimization strategies to reduce the impact of packet loss on image quality.

[0073] In step S12, it is necessary to classify the network status using a fuzzy clustering algorithm based on the initial infrared thermal imaging data and the network environment parameters, and to perform adaptive dynamic adjustment of the transmission strategy to obtain an optimized transmission strategy.

[0074] In one implementation, the fuzzy clustering algorithm is used to classify the network status according to the initial infrared thermal imaging data and the network environment parameters, and the transmission strategy is adaptively and dynamically adjusted to obtain an optimized transmission strategy, including:

[0075] According to the network environment parameters, a fuzzy clustering algorithm is used to perform a three-dimensional network status feature space analysis including signal strength, bandwidth fluctuation rate and round-trip delay, and divide the network status to obtain a real-time network status stability level; according to the initial infrared thermal imaging data, key frame distribution characteristics and temperature sensitive area analysis are performed to obtain a transmission priority weight matrix; the historical network status stability level and the historical transmission priority weight matrix are used as input, and the historical optimized transmission strategy is used as output to construct a transmission strategy adaptive dynamic adjustment model and train it. When the number of training times is greater than or equal to the preset number of training times, the training is determined to be completed, and a trained transmission strategy adaptive dynamic adjustment model is obtained; the network status stability level and the transmission priority weight matrix are input into the trained transmission strategy adaptive dynamic adjustment model to obtain an optimized transmission strategy.

[0076] It should be noted that the optimized transmission strategy refers to dynamically adjusting the data transmission mode based on the real-time network status and image data characteristics to achieve the best balance between image quality and transmission real-time under different network conditions. The acquisition of the optimized transmission strategy relies on the fuzzy clustering algorithm to classify the network status, and adaptively adjusts it in combination with the image data characteristics. First, by analyzing the network signal strength, bandwidth fluctuation rate and round-trip delay, a three-dimensional network status feature space is constructed, and the network status level is divided. Then, the key frame distribution characteristics and temperature sensitive area analysis in the infrared thermal imaging data are combined to generate a transmission priority weight matrix. Then, the historical network status stability level and the historical transmission priority weight matrix are input into the training model, and the historical optimized transmission strategy is used as the output to train the adaptive dynamic adjustment model. When the number of training times reaches the preset threshold, the model training is completed and used for real-time prediction and generation of the optimized transmission strategy. In the subsequent transmission control process, the optimized transmission strategy can dynamically adjust the data compression ratio, transmission bandwidth allocation and packet loss recovery strategy according to the current network status stability level and the transmission priority weight matrix to ensure that the key frame data is transmitted first, while taking into account the overall image quality and transmission delay, thereby improving the stability and reliability of infrared thermal imaging data transmission. In infrared thermal imaging data, key frames may contain areas with significant temperature changes, which may represent important thermal events or targets; temperature sensitive area analysis involves identifying areas in the image that are most sensitive to temperature changes, which may require higher transmission priority due to their importance in specific applications. The process of constructing the transmission priority weight matrix includes the following steps: identifying key frames and temperature sensitive areas; assigning an initial weight to each area based on the degree of temperature change and importance of the area in the application; dynamically adjusting these weights according to network conditions and transmission requirements. For example, if the network condition is poor, the weight of key frames may need to be increased to ensure that they can be successfully transmitted. The historical network status stability level and the historical transmission priority weight matrix can be obtained by continuously monitoring and collecting network performance indicators, and then using statistical methods or machine learning algorithms to analyze these data to obtain the historical network status stability level and the historical transmission priority weight matrix.

[0077] In step S13, it is necessary to use a data mining algorithm to analyze the packet loss mechanism based on the packet loss information to obtain a packet loss mapping relationship model.

[0078] In one implementation, the packet loss mechanism analysis is performed using a data mining algorithm based on the packet loss information to obtain a packet loss mapping relationship model, including:

[0079] According to the packet loss information, the data compression type of each packet loss event and the key frame identifier of the corresponding image frame are extracted to obtain a data packet attribute feature matrix; according to the packet loss information, the timestamp information of the packet loss event, the network node location identifier and the lost data packet sequence number are analyzed to obtain a spatiotemporal feature vector; according to the packet loss information, an area analysis of the temperature matrix coordinates corresponding to the lost data packet is performed, and the maximum temperature deviation value of the area is calculated to obtain a quantitative index of image quality impact; according to the data packet attribute feature matrix, the spatiotemporal feature vector and the quantitative index of image quality impact, a data mining algorithm is used to perform a packet loss mechanism analysis to obtain a dynamically adjusted packet loss mapping relationship model.

[0080] It should be noted that the packet loss mapping relationship model reflects the relationship between packet loss events and data packet characteristics, network status and image quality, so as to predict and optimize packet loss compensation strategies during transmission. The packet loss mapping relationship model acquisition relies on the in-depth analysis of the packet loss mechanism by the data mining algorithm. First, the data compression type and key frame identifier are extracted through the packet loss information to construct the data packet attribute feature matrix. At the same time, the packet loss timestamp, network node location and lost data packet sequence number are analyzed to form a spatiotemporal feature vector, and the maximum temperature deviation value of the region is calculated in combination with the temperature matrix coordinate area of ​​the lost data packet to obtain the quantitative index of image quality impact. Then, the packet loss mechanism is analyzed using the data mining algorithm to obtain a dynamically adjusted packet loss mapping relationship model. The packet loss mapping relationship model provides a basis for adaptively adjusting data redundancy strategies, such as giving priority to protecting key frames, enhancing data recovery capabilities in high temperature areas or dynamically adjusting compression ratios to reduce the impact of packet loss on infrared thermal imaging data transmission and improve data integrity and image quality. The regional maximum temperature deviation value reflects the degree of impact of packet loss events on image quality, thereby obtaining a quantitative index of image quality impact. Through data mining algorithms, such as association rule learning, cluster analysis, or machine learning models, the relationship between these features can be analyzed to identify potential causes or patterns that lead to packet loss. For example, it can be found that the packet loss rate is high at a specific network node or time period, or that certain types of data packets are more likely to be lost. These analysis results help build a packet loss mapping relationship model.

[0081] In step S14, it is necessary to dynamically balance image quality and transmission real-time performance according to the network environment parameters, the optimized transmission strategy and the packet loss mapping relationship model in combination with application scenario requirements to obtain a balance constraint condition.

[0082] In one implementation, the image quality and transmission real-time performance are dynamically balanced according to the network environment parameters, the optimized transmission strategy, and the packet loss mapping relationship model, in combination with application scenario requirements, to obtain a balance constraint condition, including:

[0083] According to the network environment parameters, the current network transmission bandwidth and the current network transmission delay are extracted, and the real-time performance of the transmission channel is combined for analysis to obtain a real-time image transmission bandwidth threshold; according to the optimized transmission strategy, the quality requirement analysis is performed in combination with a preset multi-scenario quality requirement feature library to obtain a peak signal-to-noise ratio threshold for each scenario and a maximum allowable transmission delay for each scenario; according to the image transmission bandwidth threshold, the peak signal-to-noise ratio threshold for each scenario and the maximum allowable transmission delay for each scenario, a non-dominated sorting genetic algorithm is used to perform multivariate collaborative analysis to obtain a multi-objective optimization function including a compression rate influencing factor and a delay penalty factor; according to the multi-objective optimization function and the packet loss mapping relationship model, an image quality assessment algorithm is used to perform a dynamic balance between image quality and transmission real-time performance to obtain a balance constraint condition;

[0084] Wherein, the multi-objective optimization function is as follows:

[0085]

[0086] in, represents the multi-objective optimization function, represents the image compression rate, represents the reference compression ratio, is the adjustment coefficient, is the transmission delay, is the delay penalty weight; Indicates ideal quality for each image area, Indicates The actual reception quality of each image area, represents the regional importance weight, The number of regions that the image is divided into, A coefficient for adjusting the weights of different target influences.

[0087] It should be noted that the balance constraint condition is an optimization constraint condition that coordinates image quality and transmission real-time performance in a complex network environment. The core goal is to ensure the stability and reliability of image data transmission under the conditions of limited bandwidth and changing packet loss rate, while improving transmission quality and reducing delay as much as possible. The acquisition process of the balance constraint condition includes: first, according to the network environment parameters, the current network transmission bandwidth and delay are monitored in real time, and the data transmission capacity under the current network conditions is determined by combining the optimized transmission strategy with the packet loss mapping relationship model; then, combined with the preset multi-scenario quality requirement feature library, the peak signal-to-noise ratio threshold and the maximum allowable transmission delay under different application scenarios are analyzed to ensure that the image quality meets the actual needs; then, a non-dominated sorting genetic algorithm is used to perform multi-objective optimization analysis between the image transmission bandwidth threshold, signal-to-noise ratio threshold and delay constraint, and an optimization function containing compression rate influencing factors and delay penalty factors is constructed to achieve adaptive adjustment of the transmission strategy; finally, the multi-objective optimization results are verified through the image quality evaluation algorithm, and the compression ratio, data allocation strategy and transmission priority are dynamically adjusted to finally form a balance constraint condition that meets the current network environment and application scenario requirements. Balance constraints are applied in the real-time data transmission process. By dynamically adjusting the image compression rate, network bandwidth allocation and packet loss compensation strategy, the optimal balance between image transmission quality and real-time performance is achieved, ensuring that the system can still operate stably and efficiently under complex network conditions.

[0088] In one embodiment, the image compression rate is determined based on the performance of an actual compression algorithm (such as JPEG, PNG, H.264, etc.), and the image compression rate of JPEG is generally between 10:1 and 40:1; the reference compression rate is a preset optimal value; the adjustment coefficient is determined by an optimization algorithm (such as grid search, random search, etc.), and data is collected after multiple experiments using different adjustment coefficients for regression analysis to determine the optimal value, with the goal of finding the best compression rate adjustment effect. The delay penalty weight is set by the system designer based on the needs of the application scenario. The ideal quality is defined by an image quality assessment standard (such as PSNR), while the actual reception quality is obtained by measuring the image quality after actual transmission. The regional importance weight is set based on the importance of different areas in the image. For example, in video surveillance, it may be necessary to give a higher weight to the area where the surveillance target is located.

[0089] In step S15, it is necessary to use a time series prediction algorithm to estimate future bandwidth requirements based on the balance constraint conditions, the network environment parameters and the packet loss mapping relationship model to obtain a network resource pre-allocation plan.

[0090] In one implementation, the method of estimating future bandwidth demand using a time series prediction algorithm based on the balance constraint condition, the network environment parameter and the packet loss mapping relationship model to obtain a network resource pre-allocation scheme includes:

[0091] According to the balance constraint condition and the network environment parameter, an association analysis is performed between the image compression rate gradient change rule and the transmission delay compensation strategy to obtain a bandwidth demand time series feature vector; according to the network environment parameter, combined with a preset multi-scenario quality demand feature library, a collaborative analysis of network state stability and scene quality demand is performed to obtain resource allocation weight coefficients for different transmission time periods; according to the packet loss mapping relationship model, network node position identification and packet loss event analysis are performed to obtain a channel quality assessment matrix; according to the bandwidth demand time series feature vector, the resource allocation weight coefficient, the channel quality assessment matrix and the balance constraint condition, a time series prediction algorithm is used to pre-allocate network resources to obtain a network resource pre-allocation plan.

[0092] It should be noted that the network resource pre-allocation scheme is a dynamic resource scheduling strategy based on time series prediction, which aims to optimize bandwidth allocation in advance to ensure efficient transmission of image data under different network conditions. The acquisition process of the network resource pre-allocation scheme includes: first, according to the balance constraint conditions and network environment parameters, the relationship between the image compression rate gradient change rule and the transmission delay compensation strategy is analyzed, and the time series feature vector of bandwidth demand is extracted; then, combined with the preset multi-scenario quality demand feature library, the network state stability and scene quality demand are collaboratively analyzed to determine the resource allocation weight coefficient for different transmission periods; then, based on the packet loss mapping relationship model, the historical packet loss events and packet loss distribution characteristics of the network nodes are analyzed to construct the channel quality assessment matrix; finally, the bandwidth demand trend is modeled and predicted using the time series prediction algorithm, and network resources are allocated in advance based on the prediction results to obtain the network resource pre-allocation scheme. The network resource pre-allocation scheme guides the dynamic bandwidth adjustment, priority allocation and resource scheduling strategy in the data transmission process, so that the system can still maintain the image transmission quality in a complex network environment, and minimize the impact of network congestion and data loss, thereby improving the overall transmission performance and stability. The gradient change rule of the image compression rate can be determined by monitoring the changes in image quality under different compression rates. By analyzing the timestamps of packet loss events and network node location identifiers, we can identify which network nodes or links have experienced packet loss within a specific time period. Using the packet loss information and network node location identifiers, we can evaluate the channel quality of each node or link, and then integrate the evaluated channel quality indicators into a matrix.

[0093] In step S16, it is necessary to establish a closed-loop optimization model based on the network resource pre-allocation plan and the network environment parameters in combination with transmission efficiency and cost constraints, and perform model optimization iterations to obtain a steady-state transmission plan.

[0094] In one implementation, the method of establishing a closed-loop optimization model based on the network resource pre-allocation scheme and the network environment parameters in combination with transmission efficiency and cost constraints and performing model optimization iterations to obtain a steady-state transmission scheme includes:

[0095] According to the network resource pre-allocation scheme, a multivariate collaborative analysis of bandwidth allocation gradient values, transmission priority weights, and channel quality assessment matrices is performed to obtain a multi-dimensional optimization input vector; according to the network environment parameters, combined with a preset transmission efficiency cost function, a calculation is performed to obtain a real-time transmission efficiency score and a real-time cost consumption index; according to the multi-dimensional optimization input vector, a dynamic policy gradient algorithm is used for analysis and derivation to obtain a closed-loop optimization strategy including a bandwidth adjustment coefficient, a compression ratio correction factor, and a transmission path selection parameter; according to the closed-loop optimization strategy, network resources are reallocated and the transmission strategy is iteratively updated, and real-time monitoring is performed after transmission. The image quality score, transmission delay after transmission and packet loss rate after transmission are used to obtain strategy effect feedback data; based on the strategy effect feedback data, a sliding window mechanism is used to perform spatiotemporal correlation analysis to identify the lagged effects of strategy adjustment on image quality and transmission efficiency, and obtain the strategy optimization deviation; a closed-loop optimization model is constructed based on the strategy optimization deviation, the closed-loop optimization strategy, the real-time transmission efficiency score and the real-time cost consumption index, and the closed-loop optimization model is trained. When the decrease rate of the multi-objective loss function is lower than a preset threshold in a consecutive preset number of iterations, the current optimization model parameters are determined to be a steady-state transmission scheme.

[0096] It should be noted that the steady-state transmission scheme is a long-term transmission strategy based on a closed-loop optimization model, which dynamically balances bandwidth allocation, image compression ratio adjustment, and transmission path selection in a complex network environment, thereby optimizing image quality and transmission efficiency. The process of obtaining a steady-state transmission scheme includes: first, according to the network resource pre-allocation scheme, a multi-dimensional optimization input analysis is performed in combination with the bandwidth allocation gradient value, the transmission priority weight and the channel quality assessment matrix; then, based on the network environment parameters and the preset transmission efficiency cost function, the real-time transmission efficiency score and the cost consumption index are calculated to quantify the transmission performance; then, a dynamic policy gradient algorithm is used to derive the bandwidth adjustment coefficient, the compression ratio correction factor and the transmission path selection parameter to form a preliminary closed-loop optimization strategy, and the transmission scheme is continuously optimized through network resource reallocation and iterative update of the transmission strategy; in this process, the system monitors the image quality score, transmission delay and packet loss rate after transmission in real time, obtains the policy effect feedback data, and uses the sliding window mechanism to perform spatiotemporal correlation analysis to evaluate the lag effect of the policy adjustment on the image quality and transmission efficiency, thereby calculating the policy optimization deviation; finally, the policy optimization deviation, the closed-loop optimization strategy, the real-time transmission efficiency score and the cost consumption index are used as input to construct and train the closed-loop optimization model, and when the decline rate of the multi-objective loss function is stable below the preset threshold, the optimization model parameters are determined to obtain the steady-state transmission scheme. The steady-state transmission scheme can achieve long-term stable and efficient data transmission, and maintain the best transmission effect when the network environment changes or the bandwidth fluctuates, thereby ensuring the optimal performance of the system under low-cost constraints. The sliding window mechanism is a commonly used time series analysis method that allows us to observe the changes in data over a period of time; in this process, the sliding window is used to capture the data changes before and after the policy adjustment in order to analyze the impact of the policy on image quality and transmission efficiency. Spatiotemporal correlation analysis is to identify the spatiotemporal correlation impact of policy adjustment on image quality and transmission efficiency by analyzing data at different time points and different network nodes.

[0097] In order to facilitate the understanding of the present invention, some preferred embodiments of the present invention are further described below.

[0098] The following describes the working process of the present invention using a common scenario as an example. Figure 2 , which is Figure 1 Schematic diagram of the working scenario of the method.

[0099] At a certain field environment monitoring site, a high-precision infrared thermal imaging device is installed to monitor forest fire warnings. The device needs to transmit infrared thermal imaging data to a remote monitoring center in real time so that monitoring personnel can analyze and make decisions on the fire situation. However, since the monitoring site is located in a remote area, the network environment is complex and the bandwidth is limited, problems such as packet loss, delay, and reduced transmission quality may occur during data transmission. In order to ensure that infrared thermal imaging data can be transmitted to the remote monitoring center stably and efficiently, the method of the present invention is applied to this scenario.

[0100] In this scenario, the initial infrared thermal imaging data is first acquired, while the network environment parameters and packet loss information are monitored in real time. Since the fire monitoring scenario has high requirements for data real-time, the transmission system needs to dynamically adjust to different network conditions while ensuring the quality of key images. For the acquired data, the data compression algorithm is first used to determine the maximum transmittable data volume threshold based on the current network bandwidth. If the compressed data volume exceeds the threshold, the compression ratio is further adjusted to ensure that the data can still be transmitted stably under bandwidth constraints.

[0101] Secondly, in order to improve the stability of data transmission, the fuzzy clustering algorithm is used to classify the current network status, including multiple characteristic parameters such as signal strength, bandwidth fluctuation rate and round-trip delay, to divide the network status stability level. Combined with the key frame feature analysis in infrared thermal imaging data, the importance of temperature-sensitive areas is evaluated to establish a transmission priority weight matrix. Subsequently, the adaptive dynamic adjustment model is trained using historical data, and the real-time network status level and priority weight matrix are input into the model to automatically adjust and optimize the transmission strategy to adapt to the current network environment.

[0102] During data transmission, packet loss is a key factor affecting monitoring results. The present invention analyzes the packet loss mechanism through a data mining algorithm and establishes a packet loss mapping relationship model. Specifically, the key attributes of the lost data packet are extracted, including the image frame type, temperature distribution information, timestamp, network node location, etc. to which the data packet belongs, and the specific impact of packet loss on the monitoring effect is analyzed in combination with the quantitative index of image quality impact. By dynamically adjusting the packet loss mapping relationship model, the subsequent data transmission strategy is optimized to reduce the impact of key data loss.

[0103] In the process of fire monitoring, the core goal of data transmission is to ensure the real-time performance of data while ensuring the quality of fire images. To this end, the present invention further dynamically balances image quality and transmission real-time performance through a multi-objective optimization method. Based on the real-time network bandwidth, the optimized transmission strategy and the packet loss mapping relationship, the image transmission bandwidth threshold suitable for the current network environment is calculated, and the non-dominated sorting genetic algorithm is used for collaborative optimization in combination with the peak signal-to-noise ratio requirements and the maximum allowable transmission delay of different fire monitoring scenarios. By optimizing the objective function, the image compression rate and transmission strategy are dynamically adjusted to achieve a balance between image quality and transmission real-time performance.

[0104] In addition, in order to improve the foresight of data transmission, the present invention uses a time series prediction algorithm to estimate future bandwidth requirements and formulate a network resource pre-allocation plan. By analyzing the gradient change law of image compression rate, network state stability and packet loss mapping relationship, a bandwidth demand prediction model is constructed. Combined with historical bandwidth fluctuations, the bandwidth usage trend in different time periods is estimated, so that the data transmission strategy can be adjusted in advance when the bandwidth is limited, improving the continuity and stability of transmission.

[0105] Finally, in the process of optimizing data transmission, considering the constraints of transmission efficiency and cost, the present invention establishes a closed-loop optimization model and uses a dynamic policy gradient algorithm for optimization iteration. Through parameter optimization such as bandwidth adjustment, compression ratio correction and transmission path selection, the image quality, transmission delay and packet loss rate after transmission are monitored in real time, and adjustments are made based on policy feedback data. The sliding window mechanism is used to analyze the lag effect of policy adjustment on transmission effect, and the optimization deviation is calculated to continuously improve the transmission strategy, and finally a stable adaptive data transmission solution is achieved.

[0106] In summary, the present invention discloses a signal transmission method and system for an infrared thermal imaging movement, including obtaining initial infrared thermal imaging data, network environment parameters and packet loss information; according to the initial infrared thermal imaging data and the network environment parameters, a fuzzy clustering algorithm is used to classify the network state, and a transmission strategy is adaptively and dynamically adjusted to obtain an optimized transmission strategy; according to the packet loss information, a data mining algorithm is used to analyze the packet loss mechanism to obtain a packet loss mapping relationship model; according to the network environment parameters, the optimized transmission strategy and the packet loss mapping relationship model, in combination with application scenario requirements, image quality and transmission real-time performance are dynamically balanced to obtain a balance constraint condition; according to the balance constraint condition, the network environment parameters and the packet loss mapping relationship model, a time series prediction algorithm is used to estimate future bandwidth requirements to obtain a network resource pre-allocation plan; according to the network resource pre-allocation plan and the network environment parameters, a closed-loop optimization model is established in combination with transmission efficiency and cost constraints, and model optimization iteration is performed to obtain a stable state transmission plan.

[0107] The present invention realizes efficient data transmission of infrared thermal imaging core in complex network environment by constructing adaptive dynamic adjustment signal transmission method. First, the initial infrared thermal imaging data, network environment parameters and packet loss information are obtained to provide basic data for subsequent transmission optimization. Then, in view of the dynamic change of network state, the fuzzy clustering algorithm is used to classify the network state, and the transmission strategy is adaptively and dynamically adjusted based on the classification result to obtain the optimized transmission strategy, thereby improving the adaptability of data transmission. Further, in order to deeply analyze the packet loss characteristics, a data mining algorithm is used to construct a packet loss mapping relationship model, and the packet loss mechanism is analyzed by extracting the data packet attribute characteristics, spatiotemporal characteristics and image quality impact quantitative indicators to support more refined transmission control. Next, combined with network environment parameters, optimized transmission strategy and packet loss mapping relationship model, comprehensive application scenario requirements, non-dominated sorting genetic algorithm is used to perform multi-objective optimization, dynamically balance image quality and transmission real-time, and form a balanced constraint condition. Based on the balanced constraint condition, a time series prediction algorithm is further used to predict future bandwidth requirements in combination with the bandwidth demand time series feature vector, resource allocation weight coefficient and channel quality assessment matrix, obtain a network resource pre-allocation scheme, and realize forward-looking resource allocation. Finally, based on the network resource pre-allocation scheme, combined with transmission efficiency and cost constraints, a closed-loop optimization model is constructed using a dynamic policy gradient algorithm, and optimization iterations are performed through the spatiotemporal correlation analysis of transmission feedback data. When the rate of decrease of the multi-objective loss function tends to be stable, a stable state transmission scheme is finally formed, thereby achieving long-term stable and efficient transmission. Therefore, the present invention realizes adaptive and dynamically adjusted infrared thermal imaging core data transmission, making it more stable and adaptable in complex network environments.

[0108] Reference Figure 2 The second embodiment of the present invention provides a signal transmission system for an infrared thermal imaging core, comprising:

[0109] Data acquisition module, used to obtain initial infrared thermal imaging data, network environment parameters and packet loss information;

[0110] A strategy optimization module, for classifying the network status using a fuzzy clustering algorithm according to the initial infrared thermal imaging data and the network environment parameters, and performing adaptive dynamic adjustment of the transmission strategy to obtain an optimized transmission strategy;

[0111] A packet loss analysis module, used to analyze the packet loss mechanism based on the packet loss information using a data mining algorithm to obtain a packet loss mapping relationship model;

[0112] A balance constraint module, used to dynamically balance image quality and transmission real-time performance according to the network environment parameters, the optimized transmission strategy and the packet loss mapping relationship model, combined with application scenario requirements, to obtain a balance constraint condition;

[0113] A resource allocation module, configured to estimate future bandwidth demand using a time series prediction algorithm according to the balance constraint condition, the network environment parameter and the packet loss mapping relationship model, and obtain a network resource pre-allocation plan;

[0114] The result output module is used to establish a closed-loop optimization model and perform model optimization iteration according to the network resource pre-allocation plan and the network environment parameters in combination with transmission efficiency and cost constraints to obtain a stable transmission plan.

[0115] It should be noted that the signal transmission system of an infrared thermal imaging movement provided in an embodiment of the present invention is used to execute all the process steps of a signal transmission method of an infrared thermal imaging movement in the above embodiment. The working principles and beneficial effects of the two correspond one to one, and thus will not be repeated here.

[0116] The embodiment of the present invention further provides an electronic device. The electronic device includes: a processor, a memory, and a computer program stored in the memory and executable on the processor, such as a signal transmission program for an infrared thermal imaging core. When the processor executes the computer program, the steps in the above-mentioned signal transmission method embodiments for the infrared thermal imaging core are implemented, such as Figure 1 Alternatively, when the processor executes the computer program, the functions of the modules / units in the above-mentioned device embodiments are realized, such as the resource allocation module.

[0117] Exemplarily, the computer program may be divided into one or more modules / units, which are stored in the memory and executed by the processor to implement the present invention. The one or more modules / units may be a series of computer program instruction segments capable of implementing specific functions, which are used to describe the execution process of the computer program in the electronic device.

[0118] The electronic device may be a computing device such as a desktop computer, a notebook, a PDA, and a smart tablet. The electronic device may include, but is not limited to, a processor and a memory. Those skilled in the art will appreciate that the above components are merely examples of electronic devices and do not constitute a limitation on the electronic device. The electronic device may include more or fewer components than the above components, or may combine certain components, or different components. For example, the electronic device may also include input and output devices, network access devices, buses, etc.

[0119] The processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc. The processor is the control center of the electronic device, and uses various interfaces and lines to connect various parts of the entire electronic device.

[0120] The memory can be used to store the computer program and / or module, and the processor realizes various functions of the electronic device by running or executing the computer program and / or module stored in the memory, and calling the data stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, an application required for at least one function (such as a sound playback function, an image playback function, etc.), etc.; the data storage area can store data created according to the use of the mobile phone (such as audio data, a phone book, etc.), etc. In addition, the memory can include a high-speed random access memory, and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0121] Wherein, if the module / unit integrated in the electronic device is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Wherein, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium. It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electric carrier signals and telecommunication signals.

[0122] It should be noted that the device embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. In addition, in the accompanying drawings of the device embodiments provided by the present invention, the connection relationship between the modules indicates that there is a communication connection between them, which may be specifically implemented as one or more communication buses or signal lines. A person of ordinary skill in the art may understand and implement it without paying any creative effort.

[0123] The specific embodiments described above further illustrate the purpose, technical solutions and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. It is particularly pointed out that for those skilled in the art, any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the scope of protection of the present invention.

Claims

1. A signal transmission method for an infrared thermal imaging core, characterized in that: Executed by a computer, including: Obtain initial infrared thermal imaging data, network environment parameters and packet loss information; According to the initial infrared thermal imaging data and the network environment parameters, a fuzzy clustering algorithm is used to classify the network status, and the transmission strategy is adaptively and dynamically adjusted to obtain an optimized transmission strategy; According to the packet loss information, a data mining algorithm is used to analyze the packet loss mechanism and obtain a packet loss mapping relationship model; According to the network environment parameters, the optimized transmission strategy and the packet loss mapping relationship model, combined with the application scenario requirements, dynamically balance the image quality and the real-time transmission to obtain the balance constraint condition; According to the balance constraint condition, the network environment parameter and the packet loss mapping relationship model, a time series prediction algorithm is used to estimate future bandwidth demand to obtain a network resource pre-allocation plan; According to the network resource pre-allocation scheme and the network environment parameters, a closed-loop optimization model is established in combination with transmission efficiency and cost constraints, and model optimization iterations are performed to obtain a steady-state transmission scheme.

2. The signal transmission method of the infrared thermal imaging core according to claim 1, characterized in that: The obtaining of initial infrared thermal imaging data, network environment parameters and packet loss information includes: Obtain original infrared thermal imaging data, real-time feedback of network environment parameters and packet loss information during network transmission; According to the original infrared thermal imaging data, a data compression algorithm is used in combination with a preset compression ratio threshold to perform spatial redundancy compression processing to obtain original compressed data; Calculating based on the original compressed data and the network environment parameters to obtain a maximum transmission data volume threshold corresponding to the current network bandwidth; The data volume of the original compressed data is compared with the maximum transmission data volume threshold. When the data volume of the original compressed data is greater than the maximum transmission data volume threshold, the compression ratio threshold is dynamically increased and the compression process is re-executed to obtain the initial infrared thermal imaging data; when the data volume of the original compressed data is less than the maximum transmission data volume threshold, the original compressed data is determined to be the initial infrared thermal imaging data.

3. The signal transmission method of the infrared thermal imaging core according to claim 1, characterized in that: The method of classifying the network status by using a fuzzy clustering algorithm according to the initial infrared thermal imaging data and the network environment parameters, and adaptively and dynamically adjusting the transmission strategy to obtain an optimized transmission strategy includes: According to the network environment parameters, a fuzzy clustering algorithm is used to perform a three-dimensional network status feature space analysis including signal strength, bandwidth fluctuation rate and round-trip delay and divide the network status to obtain a real-time network status stability level; According to the initial infrared thermal imaging data, key frame distribution characteristics and temperature sensitive area analysis are performed to obtain a transmission priority weight matrix; The historical network state stability level and the historical transmission priority weight matrix are used as input, and the historical optimized transmission strategy is used as output, a transmission strategy adaptive dynamic adjustment model is constructed and trained, and the training is determined to be completed when the number of training times is greater than or equal to the preset number of training times, and a transmission strategy adaptive dynamic adjustment model that has completed training is obtained; The network state stability level and the transmission priority weight matrix are input into the trained transmission strategy adaptive dynamic adjustment model to obtain an optimized transmission strategy.

4. The signal transmission method of the infrared thermal imaging core according to claim 1, characterized in that: The method of using a data mining algorithm to analyze the packet loss mechanism based on the packet loss information to obtain a packet loss mapping relationship model includes: According to the packet loss information, extract the data compression type of each packet loss event and the key frame identifier of the corresponding image frame to obtain a data packet attribute feature matrix; According to the packet loss information, analyzing the timestamp information of the packet loss event, the network node location identifier and the lost data packet sequence number to obtain a spatiotemporal feature vector; According to the packet loss information, a temperature matrix coordinate region analysis corresponding to the lost data packet is performed, and a maximum temperature deviation value of the region is calculated to obtain a quantitative index of image quality impact; According to the data packet attribute feature matrix, the spatiotemporal feature vector and the image quality impact quantitative index, a data mining algorithm is used to analyze the packet loss mechanism to obtain a dynamically adjusted packet loss mapping relationship model.

5. The signal transmission method of the infrared thermal imaging core according to claim 1, characterized in that: The method of dynamically balancing image quality and transmission real-time performance according to the network environment parameters, the optimized transmission strategy, and the packet loss mapping relationship model in combination with application scenario requirements to obtain a balance constraint condition includes: According to the network environment parameters, the current network transmission bandwidth and the current network transmission delay are extracted, and the real-time performance of the transmission channel is combined for analysis to obtain a real-time image transmission bandwidth threshold; According to the optimized transmission strategy, combined with a preset multi-scenario quality requirement feature library, quality requirement analysis is performed to obtain a peak signal-to-noise ratio threshold for each scenario and a maximum allowable transmission delay for each scenario; According to the image transmission bandwidth threshold, the peak signal-to-noise ratio threshold of each scene and the maximum allowable transmission delay of each scene, a non-dominated sorting genetic algorithm is used to perform multivariate collaborative analysis to obtain a multi-objective optimization function including a compression rate influencing factor and a delay penalty factor; According to the multi-objective optimization function and the packet loss mapping relationship model, an image quality assessment algorithm is used to dynamically balance image quality and transmission real-time performance to obtain a balance constraint condition; Wherein, the multi-objective optimization function is as follows: in, represents the multi-objective optimization function, represents the image compression rate, represents the reference compression ratio, is the adjustment coefficient, is the transmission delay, is the delay penalty weight; Indicates ideal quality for each image area, Indicates The actual reception quality of each image area, represents the regional importance weight, The number of regions that the image is divided into, A coefficient for adjusting the weights of different target influences.

6. The signal transmission method of the infrared thermal imaging core according to claim 1, characterized in that: The method of estimating future bandwidth demand by using a time series prediction algorithm according to the balance constraint condition, the network environment parameter and the packet loss mapping relationship model to obtain a network resource pre-allocation scheme includes: According to the balance constraint condition and the network environment parameter, an image compression rate gradient change rule and a transmission delay compensation strategy are analyzed to obtain a bandwidth demand time series feature vector; According to the network environment parameters, combined with a preset multi-scenario quality requirement feature library, a collaborative analysis of network state stability and scenario quality requirements is performed to obtain resource allocation weight coefficients for different transmission periods; According to the packet loss mapping relationship model, network node location identification and packet loss event analysis are performed to obtain a channel quality assessment matrix; According to the bandwidth demand timing characteristic vector, the resource allocation weight coefficient, the channel quality assessment matrix and the balance constraint condition, a time series prediction algorithm is used to pre-allocate network resources to obtain a network resource pre-allocation plan.

7. The signal transmission method of the infrared thermal imaging core according to claim 1, characterized in that: The method of establishing a closed-loop optimization model based on the network resource pre-allocation scheme and the network environment parameters in combination with transmission efficiency and cost constraints and performing model optimization iteration to obtain a stable transmission scheme includes: According to the network resource pre-allocation scheme, a multivariate collaborative analysis of bandwidth allocation gradient values, transmission priority weights, and channel quality assessment matrices is performed to obtain a multi-dimensional optimization input vector; According to the network environment parameters, combined with a preset transmission efficiency cost function, calculation is performed to obtain a real-time transmission efficiency score and a real-time cost consumption index; According to the multi-dimensional optimization input vector, a dynamic policy gradient algorithm is used to analyze and derive, and a closed-loop optimization strategy including a bandwidth adjustment coefficient, a compression ratio correction factor, and a transmission path selection parameter is obtained; According to the closed-loop optimization strategy, network resources are reallocated and transmission strategies are iteratively updated, and the image quality score after transmission, the transmission delay after transmission, and the packet loss rate after transmission are monitored in real time to obtain strategy effect feedback data; According to the feedback data of the strategy effect, a sliding window mechanism is used to perform spatiotemporal correlation analysis, identify the delayed effect of strategy adjustment on image quality and transmission efficiency, and obtain the strategy optimization deviation; A closed-loop optimization model is constructed according to the strategy optimization deviation, the closed-loop optimization strategy, the real-time transmission efficiency score and the real-time cost consumption index, and the closed-loop optimization model is trained. When the decrease rate of the multi-objective loss function is lower than a preset threshold in a consecutive preset number of iterations, the current optimization model parameters are determined to be a steady-state transmission scheme.

8. A signal transmission system for an infrared thermal imaging core, characterized in that: include: Data acquisition module, used to obtain initial infrared thermal imaging data, network environment parameters and packet loss information; A strategy optimization module, for classifying the network status using a fuzzy clustering algorithm according to the initial infrared thermal imaging data and the network environment parameters, and performing adaptive dynamic adjustment of the transmission strategy to obtain an optimized transmission strategy; A packet loss analysis module, used to analyze the packet loss mechanism based on the packet loss information using a data mining algorithm to obtain a packet loss mapping relationship model; A balance constraint module, used to dynamically balance image quality and transmission real-time performance according to the network environment parameters, the optimized transmission strategy and the packet loss mapping relationship model, combined with application scenario requirements, to obtain a balance constraint condition; A resource allocation module, configured to estimate future bandwidth demand using a time series prediction algorithm according to the balance constraint condition, the network environment parameter and the packet loss mapping relationship model, and obtain a network resource pre-allocation plan; The result output module is used to establish a closed-loop optimization model and perform model optimization iteration according to the network resource pre-allocation plan and the network environment parameters in combination with transmission efficiency and cost constraints to obtain a stable transmission plan.

9. An electronic device, characterized in that: It comprises a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, and when the processor executes the computer program, it implements the signal transmission method of the infrared thermal imaging core as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored computer program, wherein when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute the signal transmission method of the infrared thermal imaging core as described in any one of claims 1 to 7.

Images