A method and system for encrypted internet transmission of images

Abstract

This invention discloses a method and system for encrypted image transmission over the internet, relating to the fields of image processing and information security technology. The method includes: extracting image visual features and network parameters to construct a key demand index; dynamically generating a key and encrypting it in blocks, calculating an encryption efficiency index; collecting network risk data, constructing a dynamic risk field, and generating a transmission security index; outputting a comprehensive security level through a multi-objective decision engine; and based on this, calling encryption algorithms and transmission protocols to complete secure transmission, and providing feedback to optimize the system. This invention achieves dynamic adaptation of encryption strategies to image content and network status, improving transmission security and adaptability.

Description

Technical Field

[0001] This invention relates to the field of image information security and network transmission technology, and in particular to a method and system for encrypted internet transmission of images. Background Technology

[0002] In internet transmission, image data faces the risk of theft and tampering. Traditional encryption methods, often employing fixed-strength algorithms, have significant limitations. Firstly, the encryption process is disconnected from the image content and real-time network conditions, making it difficult to dynamically adjust strategies based on texture complexity and network bandwidth fluctuations, potentially leading to resource waste or security vulnerabilities. Secondly, existing technologies lack a synergistic mechanism between encryption strength, transmission efficiency, and image quality, easily resulting in transmission failures or degraded reconstruction quality in high-packet-loss or high-latency networks. With the development of 5G, 6G, and the Internet of Things, network environments and image types are becoming increasingly complex, necessitating an intelligent solution capable of dynamically adaptive encryption and transmission. Summary of the Invention

[0003] This invention proposes a method for encrypted internet transmission of images, comprising:

[0004] Extract visual features from images and network parameters to construct a key demand index;

[0005] Dynamically generate keys and encrypt them in blocks, then calculate the encryption efficiency index;

[0006] Collect network risk data, construct a dynamic risk field, and generate a transmission security index;

[0007] Based on a multi-objective decision engine, a comprehensive security level is output according to the key demand index, encryption efficiency index, and transmission security index.

[0008] Based on the overall security level, the encryption algorithm and transmission protocol are invoked to complete secure transmission, and the data is fed back to optimize the system.

[0009] The image encryption internet transmission method described above includes the following sub-steps: extracting image visual features and network parameters to construct a key demand index.

[0010] The visual feature vector of the image is extracted by a multi-level image analysis module, and network bandwidth, jitter rate and channel signal-to-noise ratio are collected simultaneously as network environment parameters.

[0011] Wavelet packet analysis is used to perform multi-resolution decomposition of image data, remove environmental noise, and extract the inherent information entropy and network-induced spectral hysteresis of the image.

[0012] Based on the weighted fusion result of information entropy and spectral lag, and combined with the real-time changing trend of network environment parameters, a key demand index is constructed.

[0013] The image encryption method for internet transmission described above includes the following sub-steps: dynamically generating a key and encrypting in blocks, and calculating the encryption efficiency index.

[0014] The encryption key is dynamically generated based on the key demand index, the image is encrypted in blocks, and the CPU usage, memory usage and encryption latency are monitored in real time during the encryption process.

[0015] Analyze the statistical correlation between encrypted blocks and calculate the difference between blocks and the entropy increase ratio.

[0016] By combining resource consumption indicators and inter-block statistical characteristics, an encryption efficiency index that reflects both computing performance and security strength is constructed.

[0017] The image encryption internet transmission method described above includes the following sub-steps: collecting network risk data, constructing a dynamic risk field, and generating a transmission security index.

[0018] Real-time collection of network packet loss rate, latency jitter data, and intrusion attack probing frequency; constructing a spatiotemporally and geographically weighted network state matrix.

[0019] A dynamic risk field is constructed by using a spatiotemporal geographic weighted regression model and combining it with a network state matrix to characterize the real-time threat distribution along the transmission path;

[0020] By coupling the dynamic risk field with the encryption efficiency index, a transmission security index is generated, which quantitatively assesses the real-time security status of the transmission channel.

[0021] The image encryption internet transmission method described above, based on a multi-objective decision engine, outputs a comprehensive security level according to the key requirement index, encryption efficiency index, and transmission security index, specifically including the following sub-steps:

[0022] The three heterogeneous indicators—key requirement index, encryption efficiency index, and transmission security index—are characterized and mapped to a unified metric space. Based on this, a multi-objective optimization function that considers security, efficiency, and resource constraints is constructed.

[0023] A fast non-dominated sorting multi-objective optimization method based on an elite strategy is used to solve the Pareto optimal solution set of the multi-objective optimization function, ensuring the diversity and convergence of the solution set;

[0024] Based on the Pareto optimal solution set, an authoritative and clearly operationally guiding characteristic comprehensive security level is generated through preset decision rules, serving as the final basis for the system to select subsequent encryption and transmission strategies.

[0025] The image encryption internet transmission method described above, which involves invoking encryption algorithms and transmission protocols based on a comprehensive security level to complete secure transmission and then feeding back to the data optimization system, specifically includes the following sub-steps:

[0026] Based on the comprehensive security level output by the multi-objective collaborative decision engine, the system automatically matches and calls the combination of asymmetric and symmetric encryption algorithms of corresponding strength from the pre-configured algorithm library, while activating the reliable transmission protocol stack corresponding to the security level.

[0027] The image data is encrypted and transmitted over the network using a matching encryption algorithm and protocol stack. After decryption is performed at the receiving end, digital signature and hash verification technology are used to verify the integrity and authenticity of the image source.

[0028] The encryption time, transmission success rate, decryption accuracy rate, and attack interception records collected during this transmission process will be used for subsequent system parameter optimization.

[0029] This invention also provides an encrypted image internet transmission system, comprising:

[0030] Image perception and key construction module: Extracts image visual features and network parameters to construct a key demand index;

[0031] Dynamic encryption and performance evaluation module: dynamically generates keys and encrypts them in blocks, and calculates the encryption efficiency index;

[0032] Transmission Risk Perception and Security Assessment Module: Collects network risk data, constructs a dynamic risk field, and generates a transmission security index;

[0033] Multi-objective decision-making and level output module: Based on the multi-objective decision-making engine, it outputs a comprehensive security level according to the key requirement index, encryption efficiency index and transmission security index;

[0034] Adaptive transmission and feedback optimization module: Based on the overall security level, it calls encryption algorithms and transmission protocols to complete secure transmission and provides feedback to optimize the system.

[0035] The present invention also provides a computer storage medium, characterized in that it comprises: at least one memory and at least one processor;

[0036] Memory, used to store one or more program instructions;

[0037] A processor for running one or more program instructions to execute any of the above-described methods for encrypted internet transmission of images.

[0038] The beneficial effects achieved by this invention are as follows:

[0039] (1) By using a multi-level image analysis module and wavelet packet analysis, image features and network parameters are accurately extracted, and a key demand index is constructed to provide a scientific basis for dynamic encryption and improve the targeting and security of encryption.

[0040] (2) Block encryption and real-time resource monitoring are adopted to calculate the encryption efficiency index, balance encryption strength and computational overhead, and avoid resource waste and transmission delay;

[0041] (3) Use a spatiotemporal geographic weighted regression model to construct a dynamic risk field, generate a transmission security index, assess network threats in real time, and enhance the anti-attack capability of the transmission process;

[0042] (4) By outputting a comprehensive security level through a multi-objective collaborative decision engine, the encryption strategy and transmission environment are adaptively matched, thereby improving the overall performance of the system.

[0043] (5) Introduce a feedback mechanism to continuously optimize model parameters, realize the system’s self-evolution and closed-loop optimization, and adapt to the ever-changing network environment and security requirements. Attached Figure Description

[0044] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0045] Figure 1 This is a flowchart of an image encryption internet transmission method provided in an embodiment of this application.

[0046] Figure 2 This is a schematic diagram of an encrypted image transmission system provided in an embodiment of this application. Detailed Implementation

[0047] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0048] Example 1

[0049] like Figure 1 As shown in the figure, an embodiment of this application provides a method for encrypted internet transmission of images, including:

[0050] Step S1: Extract image visual features and network parameters to construct the key demand index;

[0051] Based on the image to be transmitted, its visual features are extracted through a multi-level image analysis module, and real-time network environment parameters are collected simultaneously. Environmental noise is removed through wavelet packet analysis, and the inherent information entropy of the image and the network-induced spectral lag are extracted. Based on this, a key demand index is constructed.

[0052] Specifically, a multi-level image analysis module extracts the visual feature vector of the image to be transmitted, and simultaneously collects real-time network environment parameters, including network bandwidth, jitter rate, and channel signal-to-noise ratio. Based on this, wavelet packet analysis is used to perform multi-resolution decomposition of the image data, effectively removing environmental noise interference and accurately extracting the inherent information entropy reflecting the image's own characteristics, as well as the spectral lag features caused by network transmission. The information entropy and spectral lag are weighted and fused, and combined with the dynamic changing trends of network environment parameters, a key requirement index is comprehensively constructed to guide the encryption process, thus providing an adaptive control basis for subsequent image encryption and transmission. This process includes the following sub-steps:

[0053] Step S11: Extract the visual feature vector of the image through the multi-level image analysis module, and simultaneously collect network bandwidth, jitter rate and channel signal-to-noise ratio as network environment parameters.

[0054] A multi-level image analysis module extracts features from the transmitted image. This module employs a multi-scale fusion algorithm based on Gabor filter banks and wavelet transform, capturing multi-level visual features from local texture to global contours by setting filters of different directions and scales. The extracted feature maps are normalized to enhance the stability of the feature representation. After pooling compression, high-dimensional visual feature vectors are generated. Network bandwidth is obtained by monitoring the data transmission rate, jitter rate is calculated by determining the variance of the data packet arrival time interval, and the channel signal-to-noise ratio is estimated in real time using the ratio of received signal power to noise power. All environmental parameters are continuously sampled via embedded sensors or network interface cards and stored in a buffer for subsequent processing.

[0055] Step S12: Use wavelet packet analysis to perform multi-resolution decomposition on the image data, remove environmental noise, and extract the inherent information entropy of the image and the network-induced spectral hysteresis.

[0056] Wavelet packet analysis employs the Daubechies wavelet basis function to decompose image data into multiple sub-bands, covering low-frequency approximation components and high-frequency detail components. A threshold denoising algorithm is used to perform soft thresholding on high-frequency coefficients, effectively suppressing random noise and transmission interference while preserving the essential structure of the image. Based on this, the inherent information entropy of the image is calculated by performing Shannon entropy calculation on the probability distribution of the wavelet packet sub-band coefficients, quantifying the information complexity and uncertainty of the image. Simultaneously, network-induced spectral lag features are extracted, specifically by comparing the phase difference between the wavelet coefficients at the transmitting and receiving ends. Ultimately, information entropy and spectral lag serve as representations of the inherent properties of the image and the influence of the network.

[0057] Step S13: Based on the weighted fusion result of information entropy and spectral lag, and combined with the real-time changing trend of network environment parameters, construct the key demand index;

[0058] Information entropy and spectral lag are characterized to eliminate the influence of dimensions, and an adaptive weighting algorithm is used to fuse them. The weight coefficients are dynamically adjusted according to image type and network state: for highly complex images, information entropy is given higher weight; for high-latency network environments, spectral lag features are emphasized. Simultaneously, the changing trends of network bandwidth, jitter rate, and signal-to-noise ratio are monitored in real time, and the short-term mean and variance of these parameters are calculated using a sliding window to capture dynamic fluctuations in the network state. The key demand index is finally generated through a linear combination model.

[0059] Step S2: Dynamically generate the key and encrypt it in blocks, then calculate the encryption efficiency index;

[0060] Based on the key demand index, encryption keys are dynamically generated and images are encrypted in blocks. During the encryption process, the usage of computing resources is monitored in real time, and by analyzing the statistical correlation between encryption blocks, an encryption efficiency index that simultaneously represents the computational efficiency and security strength of the encryption operation is calculated.

[0061] Specifically, encryption keys are dynamically generated based on the key demand index, and block-based encryption processing is performed on the image. During the encryption process, multiple computing resource indicators, including CPU utilization, memory usage, and encryption latency, are monitored in real time. Furthermore, by analyzing the statistical correlation between each encryption block, the inter-block difference and entropy increase ratio, which reflect the encryption consistency and randomness characteristics, are calculated. Combining the above resource usage data and inter-block statistical characteristics, an encryption efficiency index is constructed that simultaneously characterizes the computational efficiency and security performance of the encryption operation. This achieves a unified quantitative assessment of resource consumption and security strength in the encryption process, specifically including the following sub-steps:

[0062] Step S21: Dynamically generate encryption keys based on the key demand index, encrypt the image in blocks, and monitor the CPU usage, memory usage and encryption latency in real time during the encryption process.

[0063] Encryption keys are dynamically generated based on a key demand index. This is calculated in real-time by evaluating multiple factors, including the sensitivity of image content, the level of environmental security threats, and historical attack patterns, ensuring that key generation meets both security requirements and adapts to dynamic scenarios. A key derivation function based on chaotic mapping or quantum random numbers is employed to adjust the key length and complexity in real-time according to changes in the key demand index, thereby enhancing the unpredictability and resistance to attacks. When encrypting images in blocks, the original image is divided into non-overlapping blocks of variable size. The block size is dynamically optimized based on image features and encryption efficiency, and lightweight or high-strength encryption algorithms are applied to encrypt each block independently to balance processing speed and security strength. During encryption, a system monitoring agent is deployed simultaneously to collect and analyze computing resource indicators such as CPU utilization, memory usage, and encryption latency in real time. By embedding performance counters and system call interfaces, the computational resource consumption of the encryption task is continuously tracked, and a timestamp mechanism is used to accurately measure the time delay from encryption initiation to completion, thereby comprehensively evaluating the real-time load and response performance of the encryption process.

[0064] Step S22: Analyze the statistical correlation between encrypted blocks and calculate the inter-block difference and entropy increase ratio;

[0065] The consistency and randomness of encryption effectiveness are evaluated by analyzing the statistical correlation between encrypted blocks. A sliding window or block matching technique is used to extract the statistical features of each encrypted block, including pixel value distribution, frequency domain coefficients, and local texture patterns. Inter-block mutual information or covariance matrices are then calculated to quantify the dependencies between blocks. Based on these features, the inter-block dissimilarity is further calculated, expressed by the following formula: ,in, Indicates the statistical difference between blocks; Indicates from The total number of combinations of selecting 2 from each encrypted block; and Indicates the sequence number of the encrypted block; and Representing encrypted blocks , The corresponding statistical feature vector; Represents statistical eigenvectors Standard deviation; Indicates encrypted block , The inter-block correlation coefficient is used to reflect the dispersion of the encrypted data by comparing the statistical distance between adjacent or random block pairs; simultaneously, the entropy increase ratio is calculated using the following formula: ,in, This represents the relative entropy increase ratio; Indicates the total number of image blocks; This represents a block index, with values ​​ranging from 1 to... ; Indicates the first Image blocks; Indicates the encrypted first Information entropy of the block; Indicates the first before encryption Information entropy of the block; Represents the maximum possible information entropy; Indicates the first The gradient norm of block entropy; The standard deviation represents the entropy; This represents the tangent function.

[0066] The randomness gain introduced by the encryption operation is evaluated by comparing the information entropy values ​​of each block before and after encryption.

[0067] Step S23: Combining resource consumption indicators and inter-block statistical characteristics, construct an encryption efficiency index that simultaneously reflects computational performance and security strength;

[0068] By combining resource consumption indicators and inter-block statistical characteristics, a comprehensive encryption efficiency index is constructed. A multi-dimensional weighted fusion model is used to quantify computational efficiency and security strength in a unified manner: the computational efficiency dimension is based on the deviation of the normalized value of the resource consumption indicator from the ideal benchmark, and weights are dynamically allocated using entropy weighting or principal component analysis to reflect the impact of system load on encryption efficiency; the security strength dimension utilizes inter-block variability and entropy increase ratio, converting them into a security score through a nonlinear mapping function, focusing on evaluating the resistance of encrypted data to statistical attacks. Finally, these two dimensions are fused using a fuzzy comprehensive evaluation method or a multi-attribute decision analysis model to generate the encryption efficiency index, expressed by the following formula: ,in, Indicates the encryption efficiency index; Indicates the statistical difference between blocks; This represents the relative entropy increase ratio; Indicates the resource load balancing factor; Indicates the prevention of zero constant; This indicates fluctuations in encryption latency; This represents the encryption delay time constant.

[0069] In addition, a historical data learning mechanism is introduced to improve the adaptability and accuracy of the index calculation model in different scenarios by iteratively optimizing the model.

[0070] Step S3: Collect network risk data, construct a dynamic risk field, and generate a transmission security index;

[0071] At the transmission end, network packet loss, latency jitter, and intrusion attack probing data are collected in real time, and a dynamic risk field is constructed using a spatiotemporal weighted regression model. This risk field is then coupled with an encryption efficiency index to generate a transmission security index, which quantitatively assesses the real-time security status of the transmission channel.

[0072] Specifically, at the transmission end, network packet loss rate, latency jitter data, and intrusion attack probing frequency are collected in real time to construct a network state matrix based on spatiotemporal geographic weighting. Furthermore, a spatiotemporal geographic weighted regression model is used in conjunction with this network state matrix to construct a dynamic risk field, accurately representing the real-time threat distribution along the transmission path. By effectively coupling the dynamic risk field with the encryption efficiency index, a transmission security index that comprehensively reflects the channel's security status is generated, thereby achieving a quantitative assessment of the real-time security status of the image data transmission channel. This process includes the following sub-steps:

[0073] Step S31: Collect network packet loss rate, latency jitter data and intrusion attack probing frequency in real time, and construct a spatiotemporally weighted network state matrix;

[0074] High-precision network probes and intrusion detection systems are deployed at the transmission end to collect network packet loss rate, latency jitter data, and intrusion attack probing frequency in real time, and time consistency of data is ensured through a time synchronization protocol. A geographic information system is used to correlate the collected data with the geographical location information of the transmission path. A spatiotemporal geographic weighting algorithm is employed to perform spatial interpolation and time series alignment of the data, constructing a multi-dimensional network state matrix. The spatiotemporal geographic weighted risk field is represented by the following formula: ,in, Indicates the location and time Risk field value; Indicates the number of observation points; Indicates the first Spatial weighting function for each observation point; These represent the weighting coefficients, used to balance the contributions of different risk factors; Indicates the first Packet loss rate at each observation point; Indicates the first Delay jitter at each observation point; Indicates the first The frequency of intrusion attack probes at each observation point;

[0075] Step S32: Using a spatiotemporal geographic weighted regression model, combined with the network state matrix, a dynamic risk field is constructed to characterize the real-time threat distribution along the transmission path;

[0076] Based on the network state matrix, a spatiotemporal geographic weighted regression model is applied to analyze the nonlinear relationship between network indicators and security threats. The regression coefficients are adjusted using a spatial distance decay function and a time decay factor to capture the non-stationarity of the data in the spatiotemporal dimension. Packet loss rate, latency jitter, and attack probing frequency are used as explanatory variables, and threat intensity is used as the dependent variable. Regression parameters for each geographic unit are calculated using locally weighted least squares fitting. To enhance real-time performance, an online learning mechanism is introduced, and an incremental update algorithm is used to adjust model parameters to adapt to rapid changes in network state. Furthermore, anomaly detection technology is combined to identify abrupt change areas in the risk field.

[0077] Step S33: Couple the dynamic risk field with the encryption efficiency index to generate a transmission security index, and quantitatively evaluate the real-time security status of the transmission channel.

[0078] To quantify the overall security status of the transmission channel, a dynamic risk field and an encryption efficiency index are coupled in multiple dimensions. The encryption efficiency index is calculated by evaluating indicators such as encryption algorithm execution speed, key update frequency, and data integrity verification success rate, reflecting the channel's proactive protection capabilities. During the coupling process, a fuzzy comprehensive evaluation method is employed, using threat intensity and the encryption efficiency index in the risk field as input variables to construct a membership function defining the security level of each variable. Through a weighted fusion algorithm, combined with expert rules and historical security event data, the weights of risk and encryption factors are dynamically allocated to generate a transmission security index. This index is output as a scalar, with a preset range of 0 to 100; a higher value indicates a more secure channel. To ensure the real-time nature of the evaluation, a feedback control mechanism is introduced to automatically adjust the coupling weights based on fluctuations in the security index. Time series analysis is used to predict the index trend. Finally, the transmission security index is output in real-time through a dashboard or API interface.

[0079] Step S4: Based on the multi-objective decision engine, output the comprehensive security level according to the key demand index, encryption efficiency index and transmission security index;

[0080] The key requirement index, encryption efficiency index, and transmission security index are input into the multi-objective collaborative decision engine. By solving the Pareto optimal solution of the three in a unified metric space, an authoritative and characteristic comprehensive security level is output as the final basis for system execution.

[0081] Specifically, the three heterogeneous indicators—key requirement index, encryption efficiency index, and transmission security index—are characterized and mapped to a unified metric space. Based on this, a multi-objective optimization function that comprehensively considers security performance, processing efficiency, and system resource constraints is constructed. Then, a fast non-dominated sorting multi-objective optimization method based on an elite strategy is used to solve this multi-objective optimization function, obtaining a Pareto optimal solution set with diversity and good convergence. Based on the obtained Pareto optimal solution set, an authoritative and clearly operationally guiding characteristic comprehensive security level is generated through preset decision rules. This level will serve as the final basis for the system to select specific encryption and transmission strategies in subsequent execution processes. The specific steps include the following:

[0082] Step S41: Characterize the three heterogeneous indicators—key requirement index, encryption efficiency index, and transmission security index—and map them to a unified metric space. Based on this, construct a multi-objective optimization function that considers security, efficiency, and resource constraints.

[0083] First, the key requirement index, encryption efficiency index, and transmission security index are standardized and preprocessed. The dimensional differences between the indices are eliminated through calculation, transforming them into dimensionless scalar values. Then, a feature mapping function is used to project these three indices into a unified multi-dimensional quantity space, ensuring comparability and synergy among the indices under the same benchmark. Subsequently, based on multiple constraints such as security performance, processing efficiency, and system resource limitations, a multi-objective optimization function is constructed. Weighting coefficients are introduced to reflect the relative importance of each objective under different application scenarios, thus forming a comprehensive mathematical optimization model.

[0084] Step S42: Use the fast non-dominated sorting multi-objective optimization method based on elitist strategy to solve the Pareto optimal solution set of the multi-objective optimization function, ensuring the diversity and convergence of the solution set;

[0085] A random population is initialized, and individuals are stratified using a fast non-dominated sorting mechanism. Non-dominated levels are assigned based on individual performance across multiple objective functions to identify potential superior solutions. Subsequently, crowding density calculations are used to evaluate the distribution density of individuals within the same non-dominated level, maintaining solution diversity and preventing premature local optima. During evolution, an elite retention strategy is introduced, directly preserving superior individuals from the parent generation to the offspring. The population is iteratively optimized through genetic operations such as selection, crossover, and mutation, and a fitness function is used to evaluate the overall performance of each solution until a predefined termination condition is met.

[0086] Step S43: Based on the Pareto optimal solution set, generate an authoritative, characteristic comprehensive security level with clear operational guidance through preset decision rules, which serves as the final basis for the system to select subsequent encryption and transmission strategies;

[0087] Cluster analysis is performed on the Pareto optimal solution set to identify the main distribution patterns and representative solutions. A multi-attribute decision method is then used to comprehensively evaluate each solution, considering dimensions such as security strength, efficiency, and resource consumption. Subsequently, based on pre-defined decision rules, the most representative optimal solution is selected from the solution set and mapped to a characteristic comprehensive security level system. This system is typically divided into multiple discrete levels, such as "extremely high security," "high security," and "medium security," with each level assigned specific encryption algorithm recommendations, transmission protocol configurations, and resource allocation strategies.

[0088] Step S5: Based on the overall security level, invoke the encryption algorithm and transmission protocol to complete secure transmission, and provide feedback to optimize the system.

[0089] Based on the overall security level, the system automatically invokes the corresponding strong encryption algorithm library and transmission protocol stack to complete image encryption and transmission. After decryption and integrity verification at the receiving end, the performance data of this transmission is fed back to S1 for continuous optimization of model parameters, enabling the system's self-evolution and closed-loop optimization.

[0090] Specifically, based on the comprehensive security level determined by the multi-objective collaborative decision engine, the system automatically matches and calls a combination of asymmetric and symmetric encryption algorithms corresponding to the security level from the pre-set algorithm library, while activating the adapted reliable transmission protocol stack to complete the encryption and network transmission of image data. After data decryption at the receiving end, the system further uses digital signature and hash verification technology to perform dual verification of the integrity and authenticity of the image content. On this basis, the system collects multi-dimensional performance data, including encryption time, transmission success rate, decryption accuracy, and attack interception status, and provides feedback. By constructing a historical database and establishing a parameter prediction model based on a time series prediction model, the system can mine hidden patterns and long-term dependencies from historical parameter adjustment sequences, thereby predicting the key parameter combinations for the next stage and accelerating model convergence. Further combined with the feedback adjustment mechanism algorithm, the system can smoothly and adaptively incrementally correct the weights of the feature extraction network and the core parameters of the key generation algorithm based on the deviation between the prediction results and real-time feedback, thereby realizing the system's continuous self-evolution and closed-loop optimization capabilities in a dynamic network environment. This includes the following sub-steps:

[0091] Step S51: Based on the comprehensive security level output by the multi-objective collaborative decision engine, automatically match and call the combination of asymmetric encryption algorithm and symmetric encryption algorithm of corresponding strength from the pre-configured algorithm library, and activate the reliable transmission protocol stack corresponding to the security level.

[0092] Asymmetric encryption algorithms can be designed based on elliptic curve cryptography or the large integer factorization problem, while symmetric encryption algorithms employ block ciphers or stream ciphers, dynamically adjusting key length, operating mode, and initialization vector according to the security level. Protocol stack activation involves transport layer security protocols or customized secure channel protocols. By monitoring network latency, bandwidth fluctuations, and threat intelligence in real time, it adaptively selects protocol versions and cipher suites to ensure a balance between transmission efficiency and security.

[0093] Step S52: Use a matching encryption algorithm and protocol stack to encrypt and transmit the image data over the network. After decryption at the receiving end, use digital signature and hash verification technology to verify the integrity and authenticity of the image source.

[0094] At the sending end, the image data is preprocessed using the invoked encryption algorithm, including block segmentation, padding, and encoding. Then, a symmetric encryption key is calculated and securely transmitted to the receiving end using an asymmetric encryption algorithm. After encryption, the image data is encapsulated into protocol data units and transmitted over the network via the activated transport protocol stack. During this transmission, the protocol stack implements congestion control and error retransmission to ensure reliable transmission.

[0095] At the receiving end, decryption is first performed to restore the original image data. Then, an integrity verification process is initiated: the digital signature is extracted, and a hash value is calculated using a formula. This hash value is compared with the signature and hash benchmark attached to the transmitted data to verify whether the data has been tampered with and whether its source is trustworthy. The verification process incorporates timestamps and sequence numbers to prevent replay and forgery attacks. Finally, a verification report is generated and recorded in the audit log.

[0096] Step S53: Feed back the performance data collected during this transmission process, such as encryption time, transmission success rate, decryption accuracy rate, and attack interception records, to S1 for subsequent system parameter optimization.

[0097] Step S531: Systematically collect and organize multi-dimensional performance data after each transmission task, and establish a historical database that includes encryption performance, transmission reliability and security defense effectiveness;

[0098] After each transmission task is completed, a data collection process is automatically triggered, capturing raw performance metrics such as encryption time, transmission success rate, decryption accuracy, and attack interception records through distributed sensors and log interfaces. The collected data is cleaned, denoised, and normalized, then organized by task identifier and time series, and stored in a structured historical database.

[0099] Step S532: Construct a parameter prediction model based on the time series prediction model. By analyzing the implicit patterns and long-term dependencies in the historical parameter adjustment series, predict the key parameter combinations in the next optimization cycle to accelerate the model convergence process.

[0100] By utilizing parameter adjustment sequences and performance data stored in a historical database, a parameter prediction model based on a time series prediction model is constructed. Taking historical parameter combinations and corresponding performance indicators as input, a multi-layer LSTM unit captures long-term dependencies and nonlinear patterns in the time series, outputting predicted values ​​of key parameters for the next optimization cycle. During training, gradient descent and backpropagation algorithms are used to optimize network weights, and an attention mechanism is introduced to enhance the ability to focus on key historical events, thereby improving prediction accuracy and convergence speed.

[0101] Step S533: Using a feedback adjustment mechanism algorithm, based on the deviation between the prediction result and the real-time feedback, the core parameters of the feature extraction network weight and the key generation algorithm are smoothly and adaptively incrementally corrected, thereby realizing the system's self-evolution and closed-loop optimization in a dynamic environment.

[0102] The parameters output by the prediction model and the performance deviation in real time are input into an adaptive controller based on the proportional-integral-derivative principle. The parameter correction amount is dynamically obtained based on the deviation, and the weights of the feature extraction algorithm and the core parameters of the key generation algorithm are smoothly and adaptively incrementally corrected.

[0103] Example 2

[0104] like Figure 2 As shown, Embodiment 2 of this application provides an encrypted image internet transmission system, comprising:

[0105] Image perception and key construction module 21: dynamically generates keys and encrypts them in blocks, and calculates the encryption efficiency index; it includes the following sub-modules:

[0106] Feature and parameter perception submodule 211: Extracts visual feature vectors of images through a multi-level image analysis module, and simultaneously collects network bandwidth, jitter rate and channel signal-to-noise ratio as network environment parameters;

[0107] Wavelet feature analysis submodule 212: Uses wavelet packet analysis to perform multi-resolution decomposition of image data, removes environmental noise, and extracts the inherent information entropy of the image and the network-induced spectral hysteresis;

[0108] Key demand generation submodule 213: Based on the weighted fusion result of information entropy and spectral lag, and combined with the real-time changing trend of network environment parameters, a key demand index is constructed;

[0109] Dynamic Encryption and Performance Evaluation Module 22: Dynamically generates keys and performs block encryption, calculating the encryption efficiency index; includes the following sub-modules:

[0110] Dynamic encryption execution submodule 221: dynamically generates encryption keys based on the key demand index, performs block encryption on the image, and monitors CPU usage, memory usage and encryption latency in real time during the encryption process;

[0111] Security Feature Analysis Submodule 222: Analyzes the statistical correlation between encrypted blocks and calculates the inter-block difference and entropy increase ratio;

[0112] Encryption performance evaluation submodule 223: Combining resource consumption indicators and inter-block statistical characteristics, construct an encryption efficiency index that simultaneously reflects computational performance and security strength;

[0113] Transmission Risk Awareness and Security Assessment Module 23: Collects network risk data, constructs a dynamic risk field, and generates a transmission security index; includes the following sub-modules:

[0114] Risk data acquisition submodule 231: Real-time acquisition of network packet loss rate, latency jitter data and intrusion attack probing frequency, and construction of a spatiotemporally and geographically weighted network state matrix;

[0115] Dynamic Risk Field Construction Submodule 232: Utilizes a spatiotemporal geographic weighted regression model and combines it with the network state matrix to construct a dynamic risk field that characterizes the real-time threat distribution along the transmission path;

[0116] Security Index Generation Submodule 233: Couples the dynamic risk field with the encryption efficiency index to generate a transmission security index, which quantitatively evaluates the real-time security status of the transmission channel;

[0117] Multi-objective decision-making and level output module 24: Based on a multi-objective decision-making engine, it outputs a comprehensive security level according to the key requirement index, encryption efficiency index, and transmission security index; it includes the following sub-modules:

[0118] Indicator Fusion and Modeling Submodule 241: Characterizes and maps the three heterogeneous indicators—key requirement index, encryption efficiency index, and transmission security index—to a unified metric space, and on this basis, constructs a multi-objective optimization function that considers security, efficiency, and resource constraints;

[0119] Optimization solution submodule 242: The Pareto optimal solution set of the multi-objective optimization function is solved by using a fast non-dominated sorting multi-objective optimization method based on elitist strategy, ensuring the diversity and convergence of the solution set;

[0120] Level Decision Submodule 243: Based on the Pareto optimal solution set, it generates an authoritative, characteristic comprehensive security level with clear operational guidance significance through preset decision rules, which serves as the final basis for the system to select subsequent encryption and transmission strategies;

[0121] Adaptive Transmission and Feedback Optimization Module 25: Based on the overall security level, it invokes encryption algorithms and transmission protocols to complete secure transmission and provides feedback to optimize the data system; it includes the following sub-modules:

[0122] Strategy scheduling submodule 251: Based on the comprehensive security level output by the multi-objective collaborative decision engine, automatically match and call the combination of asymmetric encryption algorithm and symmetric encryption algorithm of corresponding strength from the pre-configured algorithm library, and activate the reliable transmission protocol stack corresponding to the security level.

[0123] Secure transmission submodule 252: Uses a matching encryption algorithm and protocol stack to encrypt and transmit image data over the network. After decryption at the receiving end, it uses digital signature and hash verification technology to verify the integrity and authenticity of the image source.

[0124] Feedback collection submodule 253: Feeds back the performance data collected during this transmission process, such as encryption time, transmission success rate, decryption accuracy rate, and attack interception records, to S1 for subsequent system parameter optimization.

[0125] Corresponding to the above embodiments, the present invention provides a computer storage medium, including: at least one memory and at least one processor;

[0126] The memory is used to store one or more program instructions;

[0127] A processor is used to run one or more program instructions to execute a method for encrypted internet transmission of images.

[0128] Corresponding to the above embodiments, this embodiment of the invention provides a computer-readable storage medium containing one or more program instructions, which are executed by a processor to provide a method for encrypted internet transmission of images.

[0129] The embodiments disclosed in this invention provide a computer-readable storage medium storing computer program instructions, which, when executed on a computer, cause the computer to perform the aforementioned method for encrypted image transmission over the Internet.

[0130] In this embodiment of the invention, the processor can be an integrated circuit chip with signal processing capabilities. The processor can be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0131] The various methods, steps, and logic diagrams disclosed in the embodiments of this invention can be implemented or executed. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the methods disclosed in the embodiments of this invention can be directly embodied in the execution of a hardware decoding processor, or executed by a combination of hardware and software modules in the decoding processor. The software modules can reside in random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, or other mature storage media in the art. The processor reads information from the storage medium and, in conjunction with its hardware, completes the steps of the above methods.

[0132] The storage medium can be memory, such as volatile memory or non-volatile memory, or may include both volatile and non-volatile memory.

[0133] Among them, non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory.

[0134] Volatile memory can be random access memory (RAM), which is used as an external cache. By way of example, but not limitation, many forms of RAM are available, such as static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDRSDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous linked dynamic random access memory (Synchlink DRAM, SLDRAM), and direct memory bus RAM (DRRAM).

[0135] The storage media described in the embodiments of the present invention are intended to include, but are not limited to, these and any other suitable types of memory.

[0136] Those skilled in the art will recognize that, in one or more of the examples above, the functions described in this invention can be implemented using a combination of hardware and software. When applied as software, the corresponding functions can be stored in a computer-readable medium or transmitted as one or more instructions or code on a computer-readable medium. Computer-readable media include computer storage media and communication media, wherein communication media include any medium that facilitates the transmission of computer programs from one place to another. Storage media can be any available medium that can be accessed by a general-purpose or special-purpose computer.

[0137] The specific embodiments described above further illustrate the purpose, technical solution, 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. Any modifications, equivalent substitutions, or improvements made on the basis of the technical solution of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for encrypted internet transmission of images, characterized in that, include: Extract visual features from images and network parameters to construct a key demand index; Dynamically generate keys and encrypt them in blocks, then calculate the encryption efficiency index; Collect network risk data, construct a dynamic risk field, and generate a transmission security index; Based on a multi-objective decision engine, a comprehensive security level is output according to the key demand index, encryption efficiency index, and transmission security index. Based on the overall security level, the encryption algorithm and transmission protocol are invoked to complete secure transmission, and the data is fed back to optimize the system. Extracting visual features from images and network parameters to construct a key demand index includes the following sub-steps: The visual feature vector of the image is extracted by a multi-level image analysis module, and network bandwidth, jitter rate and channel signal-to-noise ratio are collected simultaneously as network parameters. Wavelet packet analysis is used to perform multi-resolution decomposition of image data, remove environmental noise, and extract the inherent information entropy and network-induced spectral hysteresis of the image. Based on the weighted fusion result of information entropy and spectral lag, and combined with the real-time changing trend of network parameters, a key demand index is constructed. Dynamically generate keys and encrypt them in blocks, then calculate the encryption efficiency index. This process includes the following sub-steps: The encryption key is dynamically generated based on the key demand index. The image is encrypted in blocks, and the CPU usage, memory usage and encryption latency are monitored in real time as indicators of resource consumption. Analyze the statistical correlation between encrypted blocks and calculate the inter-block difference and entropy increase ratio as statistical characteristics between blocks; By combining resource consumption indicators and inter-block statistical characteristics, an encryption efficiency index that reflects both computing performance and security strength is constructed. Collecting network risk data, constructing a dynamic risk field, and generating a transmission security index includes the following sub-steps: Real-time collection of network packet loss rate, latency jitter data, and intrusion attack probing frequency; constructing a spatiotemporally and geographically weighted network state matrix. A dynamic risk field is constructed by using a spatiotemporal geographic weighted regression model and combining it with a network state matrix to characterize the real-time threat distribution along the transmission path; By coupling the dynamic risk field with the encryption efficiency index, a transmission security index is generated, which quantitatively assesses the real-time security status of the transmission channel.

2. The method for encrypted internet transmission of images according to claim 1, characterized in that, Based on a multi-objective decision engine, a comprehensive security level is output according to the key demand index, encryption efficiency index, and transmission security index, specifically including the following sub-steps: The three heterogeneous indicators—key requirement index, encryption efficiency index, and transmission security index—are characterized and mapped to a unified metric space. Based on this, a multi-objective optimization function that considers security, efficiency, and resource constraints is constructed. A fast non-dominated sorting multi-objective optimization method based on an elite strategy is used to solve the Pareto optimal solution set of the multi-objective optimization function, ensuring the diversity and convergence of the solution set; Based on the Pareto optimal solution set, a characteristic comprehensive security level is generated through preset decision rules, which serves as the final basis for the system to select subsequent encryption and transmission strategies.

3. The method for encrypted internet transmission of images according to claim 1, characterized in that, Based on the overall security level, the encryption algorithm and transmission protocol are invoked to complete secure transmission, and the data is fed back to optimize the system. This process includes the following sub-steps: Based on the comprehensive security level output by the multi-objective decision engine, the system automatically matches and calls the combination of asymmetric and symmetric encryption algorithms of corresponding strength from the pre-configured algorithm library, while activating the reliable transmission protocol stack corresponding to the security level. The image data is encrypted and transmitted over the network using a matching encryption algorithm and protocol stack. After decryption is performed at the receiving end, digital signature and hash verification technology are used to verify the integrity and authenticity of the image source. The encryption time, transmission success rate, decryption accuracy rate, and attack interception records collected during this transmission process will be used for subsequent system parameter optimization.

4. A system for encrypted internet transmission of images, comprising executing a method for encrypted internet transmission of images as described in any one of claims 1-3, characterized in that, include: Image perception and key construction module: Extracts image visual features and network parameters to construct a key demand index; Dynamic encryption and performance evaluation module: dynamically generates keys and encrypts them in blocks, and calculates the encryption efficiency index; Transmission Risk Perception and Security Assessment Module: Collects network risk data, constructs a dynamic risk field, and generates a transmission security index; Multi-objective decision-making and level output module: Based on the multi-objective decision-making engine, it outputs a comprehensive security level according to the key requirement index, encryption efficiency index and transmission security index; Adaptive transmission and feedback optimization module: Based on the overall security level, it calls encryption algorithms and transmission protocols to complete secure transmission and provides feedback to optimize the system.

5. A computer storage medium, characterized in that, include: At least one memory and at least one processor; Memory, used to store one or more program instructions; A processor for running one or more program instructions to perform a method for encrypted internet transmission of images as described in any one of claims 1-3.

Images