Industrial and trade enterprise security risk intelligent management system and method

Abstract

The invention discloses an industry and trade enterprise security risk intelligent management system and method, and relates to the technical field of information security. The method comprises the following steps: firstly, obtaining a preprocessed industrial and trade enterprise original image, and segmenting to obtain a sub-original image set; updating a preset chaotic mapping parameter according to historical data to obtain an optimized parameter, and inputting the optimized parameter into mapping iteration to generate a chaotic sequence set; then, scrambling and diffusion control keys are extracted from the sequence according to a preset rule, and the two keys are used for performing data conversion and encryption on the target original sub-image to obtain an encrypted sub-image; and then, according to a preset rule, storing the sub-encrypted images in different physical positions, and finally, through a preset transmission path, transmitting a storage path and the two secret keys to a cloud end. According to the method, parameters are optimized according to historical data to improve the key suitability and security, the complexity and cracking resistance are enhanced through chaotic encryption, the leakage risk is reduced through sub-image scattered storage, and the whole-process security of enterprise production monitoring image data is comprehensively guaranteed.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of information security, and particularly relates to a security risk intelligent management system and method for industrial and trade enterprises. BACKGROUND

[0002] With the continuous expansion of the production scale of industrial and trade enterprises and the increasing emphasis on safety production, the intelligent management of safety risks in industrial and trade enterprises has become a key link to ensure the smooth operation of enterprises and avoid safety accidents. However, the current intelligent management of safety risks in industrial and trade enterprises faces many real challenges, such as the complex production processes and numerous equipment and facilities involved in industrial and trade enterprises, and the diverse types of safety risks that need to be controlled. These situations not only result in incomplete identification of safety risks and control blind spots, but also may cause problems such as untimely discovery of safety hazards and difficulty in timely suppressing accident clues. Therefore, it is an urgent need to build an efficient and intelligent safety risk intelligent management system for industrial and trade enterprises and to use advanced technology to achieve comprehensive, accurate, and dynamic control of all safety elements in industrial and trade enterprises to ensure the efficient promotion of safety production in industrial and trade enterprises. The traditional safety risk management method for industrial and trade enterprises has the problems of poor encryption effect due to fixed chaotic mapping parameters without dynamic optimization, concentrated storage of encrypted images, and unprogrammed transmission path that is prone to data loss or leakage, and imperfect data conversion and encryption logic, making it difficult to achieve efficient and intelligent control.

[0003] Patent CN118886042A discloses a method for encrypting and storing enterprise private data. This method converts the enterprise private data to be encrypted into a binary sequence and then segments it. Each segment of binary data is converted into a two-dimensional matrix through a binary conversion method. The classification categories are set by the values of the classification indicators of the data points in the two-dimensional matrix, and all data points are classified to generate a result matrix with the same number of classifications as the number of data points, each of which has been subjected to binary processing. The high-light points in the result matrix are determined in the Radon parameter space by performing Radon transformation on each result matrix. The position of the high-light points is scrambled to achieve the scrambling of the binary data points in the result matrix. Thus, the scrambling of each data point in the two-dimensional matrix formed by the enterprise private data is achieved. Each scrambled data point in the two-dimensional matrix is converted into a binary sequence segment and spliced in a specified order to obtain the encrypted ciphertext of the enterprise private data, achieving efficient and secure encryption of the enterprise private data. However, this invention does not consider the relevance of industrial and trade safety images and risk elements, lacks image optimization segmentation, key dynamic optimization, and storage and transmission planning, and does not achieve the coordination of encryption and control, thus resulting in low efficiency of industrial and trade enterprise safety intelligent management. SUMMARY

[0004] The purpose of the present application is to solve the problem of low efficiency of industrial and trade enterprise safety intelligent management, and to propose a security risk intelligent management system and method for industrial and trade enterprises.

[0005] In the first aspect of the embodiment of the present application, an intelligent safety risk management method for industrial and trade enterprises is first proposed, which comprises: obtaining an original image of an industrial and trade enterprise; the original image is a pre-processed enterprise production monitoring image; segmenting the original image to obtain a set of sub-original images; updating the parameters of a preset chaotic mapping according to historical data to obtain optimized parameters; inputting the optimized parameters into the preset chaotic mapping for iteration to obtain a set of chaotic sequences; obtaining a scrambling control key and a diffusion control key according to a preset combination rule for the chaotic sequences in the set of chaotic sequences; performing data conversion and encryption on a target sub-original image according to the scrambling control key and the diffusion control key to obtain a sub-encrypted image; the target sub-original image is any one of the set of sub-original images; storing the sub-encrypted image in different physical locations according to a preset storage rule; transmitting the sub-encrypted image storage path, the scrambling control key and the diffusion control key to the cloud through a preset transmission path.

[0006] Optionally, the original image is preliminarily divided to obtain a set of preliminary sub-original images; calculating the information redundancy of the preliminary sub-images in the set of preliminary sub-images to obtain sub-image information redundancy; calculating the quantum correlation degree between each preliminary sub-image in the set of preliminary sub-images to obtain sub-image correlation degree; adjusting the initial division parameters according to the sub-image information redundancy and the sub-image correlation degree to obtain target division parameters; dividing the original image according to the target division parameters to obtain the set of sub-original images.

[0007] Optionally, the resolution and information entropy of the original image are extracted to obtain an image feature vector; quantum analog processing is performed on the original image to obtain a quantum feature vector; inputting the image feature vector and the quantum feature vector into a preset neural network model of a fusion quantum computing module to obtain initial division parameters; preliminarily dividing the original image according to the initial division parameters to obtain a set of preliminary sub-images.

[0008] Optionally, a candidate parameter set of the chaotic optimization algorithm and an adaptive function are constructed; the candidate parameters in the candidate parameter set are generated by a random function within a preset value range; and the adaptive function is established according to information entropy and pixel correlation indexes of an image. A target candidate parameter is input into a preset chaotic mapping to obtain a chaotic sequence; the target candidate parameter is any one of the candidate parameters in the candidate parameter set. A historical original image is encrypted according to a historical scrambling control key, a historical diffusion control key and the chaotic sequence to obtain a historical encrypted image. A fitness value is calculated according to the adaptive function and the historical encrypted image to obtain a candidate parameter fitness value. A candidate parameter in the candidate parameter set corresponding to the maximum candidate parameter fitness value is selected as an optimized parameter. Optionally, a one-dimensional sequence is obtained by performing binary conversion on the target sub-original image. Firstly, the one-dimensional sequence is scrambled to obtain a scrambled one-dimensional sequence. Secondly, the scrambled one-dimensional sequence is diffused to obtain a scrambled sub-image. The one-dimensional sequence is scrambled to obtain a scrambled one-dimensional sequence; wherein, is a next state obtained by transforming the i th element in the one-dimensional sequence, is a binary value of a previous element of the i th element in the one-dimensional sequence, is a binary value of a next element of the i th element in the one-dimensional sequence, is a binary value of the i th element in the one-dimensional sequence, is a binary value of the i th element in the one-dimensional sequence, is a binary value of the i th element in the one-dimensional sequence, is a binary value of the i th element in the one-dimensional sequence, is a binary value of the i th element in the one-dimensional sequence, is a binary value of the i th element in the one-dimensional sequence, is an exclusive or operator symbol; and the scrambled one-dimensional sequence is converted back to a grayscale value array to obtain a scrambled sub-image. The scrambled image is diffused in a forward and reverse direction according to the diffusion control key to obtain a sub-encrypted image.

[0009] In a second aspect of the embodiment of the present application, an intelligent safety risk management system for a trade enterprise is provided, comprising: An original image acquisition module is configured to acquire an original image of the trade enterprise; the original image is a preprocessed enterprise production monitoring image; A segmentation module is configured to segment the original image to obtain a set of sub-original images; A parameter updating module is configured to update parameters of a preset chaotic mapping according to historical data to obtain optimized parameters; A chaotic sequence generation module is configured to input the optimized parameters into the preset chaotic mapping for iteration to obtain a set of chaotic sequences; The key generation module is configured to obtain a scrambling control key and a diffusion control key from a chaotic sequence in the chaotic sequence set according to a preset combination rule. The sub-encryption image generation module is configured to perform data conversion and encryption on a target sub-original image to obtain a sub-encryption image according to the scrambling control key and the diffusion control key; the target sub-original image is any one of the sub-original image set. The storage module is configured to store the sub-encryption image in different physical locations according to a preset storage rule. The transmission module is configured to transmit the sub-encryption image storage path, the scrambling control key and the diffusion control key to the cloud through a preset transmission path.

[0010] Optionally, the segmentation module comprises: The preliminary sub-original image generation module is configured to perform preliminary division on the original image to obtain a preliminary sub-original image set. The redundancy calculation module is configured to calculate information redundancy of a preliminary sub-image in the preliminary sub-image set to obtain sub-image information redundancy. The correlation degree calculation module is configured to calculate quantum correlation degree between each preliminary sub-image in the preliminary sub-image set to obtain sub-image correlation degree. The target division parameter generation module is configured to adjust an initial division parameter according to the sub-image information redundancy and the sub-image correlation degree to obtain a target division parameter. The division module is configured to divide the original image according to the target division parameter to obtain the sub-original image set.

[0011] Optionally, the preliminary sub-original image generation module comprises: The image feature extraction module is configured to extract resolution and information entropy of the original image to obtain an image feature vector. The quantum feature vector generation module is configured to perform quantum simulation processing on the original image to obtain a quantum feature vector. The initial division parameter generation module is configured to input the image feature vector and the quantum feature vector into a preset neural network model of a fusion quantum calculation module to obtain an initial division parameter. The preliminary division module is configured to preliminarily divide the original image according to the initial division parameter to obtain a preliminary sub-image set.

[0012] Optionally, the parameter update module comprises: The parameter function construction module is configured to construct a candidate parameter set and an adaptability function of the chaotic optimization algorithm; a candidate parameter in the candidate parameter set is obtained by a random function in a preset value range; the adaptability function is established according to information entropy and pixel correlation index of an image. a chaos sequence generation module configured to input a target candidate parameter into a preset chaos mapping to obtain a chaos sequence, the target candidate parameter being any one of the candidate parameters in the candidate parameter set; a historical encrypted image generation module configured to encrypt a historical original image according to a historical scrambling control key, a historical diffusion control key and the chaos sequence to obtain a historical encrypted image; a fitness value calculation module configured to calculate a fitness value of the historical encrypted image according to the fitness function to obtain a candidate parameter fitness value; an optimized parameter generation module configured to select a candidate parameter in the candidate parameter set corresponding to the maximum candidate parameter fitness value as an optimized parameter.

[0013] Optionally, the sub-encrypted image generation module comprises: a one-dimensional sequence generation module configured to perform binary conversion on the target sub-original image to obtain a one-dimensional sequence; a scrambling operation module configured to first perform scrambling on the one-dimensional sequence by , and then by , wherein, is a next state obtained after transformation of an element in the one-dimensional sequence, is a binary value of a previous element of the element in the one-dimensional sequence, is a binary value of the previous element of the element in the one-dimensional sequence, is a binary value of a next element of the element in the one-dimensional sequence, is the binary value of the next element of the element in the one-dimensional sequence, is a binary value of the element in the one-dimensional sequence, is a binary value of the element in the one-dimensional sequence, is a binary value of the element in the one-dimensional sequence, is a binary value of an element in the scrambling control key, is the binary value of the element in the scrambling control key, is an exclusive or operator symbol; and a scrambled sub-image generation module configured to convert the scrambled one-dimensional sequence back to a gray value array to obtain a scrambled sub-image. a sub-encrypted image generation module configured to perform forward and reverse diffusion on the scrambled image according to the diffusion control key to obtain a sub-encrypted image.

[0014] The beneficial effects of this invention are as follows: This invention proposes an intelligent management method for security risks in industrial and trading enterprises. First, it acquires pre-processed original images of the industrial and trading enterprise and segments them to obtain sub-original image sets. Then, it updates preset chaotic mapping parameters based on historical data, obtains optimized parameters, and inputs them into the mapping iteration to generate a chaotic sequence set. Subsequently, it extracts scrambling and diffusion control keys from the sequences according to preset rules, uses these two keys to perform data transformation and encryption on the target sub-original images, and obtains sub-encrypted images. Then, according to preset rules, the sub-encrypted images are stored in different physical locations, and finally, the storage paths and the two keys are transmitted to the cloud via preset transmission paths. This method improves key adaptability and security by optimizing parameters based on historical data, enhances complexity and anti-cracking capabilities through chaotic encryption, and reduces the risk of leakage through distributed storage of sub-images, comprehensively ensuring the security of enterprise production monitoring image data throughout the entire process. Attached Figure Description

[0015] The invention will now be further described with reference to the accompanying drawings.

[0016] Figure 1 A flowchart illustrating an intelligent safety risk management method for industrial and commercial enterprises, provided as an embodiment of the present invention; Figure 2 This is a framework diagram of an intelligent management system for safety risks in industrial and commercial enterprises, provided as an embodiment of the present invention. Detailed Implementation

[0017] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0018] 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.

[0019] This invention provides an intelligent management method for safety risks in industrial and commercial enterprises. See also... Figure 1 , Figure 1 A flowchart illustrating an intelligent safety risk management method for industrial and commercial enterprises, provided as an embodiment of the present invention.

[0020] The method includes the following steps: S101, Obtain the original image of the industrial and trading enterprise; S102, Segment the original image to obtain a sub-original image set; S103, Update the parameters of the preset chaotic mapping based on historical data to obtain optimized parameters; S104, input the optimization parameter to the preset chaotic mapping for iteration to obtain a chaotic sequence set; S105, obtain a scrambling control key and a diffusion control key according to the chaotic sequence in the chaotic sequence set according to a preset combination rule; S106, perform data conversion and encryption on the target sub-original image according to the scrambling control key and the diffusion control key to obtain a sub-encrypted image; S107, store the sub-encrypted image in different physical positions according to a preset storage rule; S108, transmit the storage path of the sub-encrypted image, the scrambling control key and the diffusion control key to the cloud through a preset transmission path; The original image is a preprocessed enterprise production monitoring image. The target sub-original image is any one of the sub-original image set.

[0021] The method provided by the embodiment of the application is used for intelligently managing the security risk of a trade enterprise, and comprises the following steps: first, obtaining a preprocessed original image of the trade enterprise, and segmenting the original image to obtain a sub-original image set; then, updating the parameters of a preset chaotic mapping according to historical data, inputting the optimized parameters into the mapping for iteration, and generating a chaotic sequence set; subsequently, extracting a scrambling control key and a diffusion control key from the sequence according to a preset rule, performing data conversion and encryption on a target sub-original image by using the two kinds of keys, and obtaining a sub-encrypted image; then, storing the sub-encrypted image in different physical positions according to a preset rule; finally, transmitting the storage path and the two kinds of keys to the cloud through a preset transmission path.

[0022] In an implementation, the image of enterprise production monitoring refers to initial image data directly collected from camera equipment at key monitoring points such as production workshops, storage areas, and pipeline stations, and is formed after a series of targeted technical processing, and is more in line with the needs of subsequent analysis and application. The core value of such original images lies in the real recording of key information such as personnel operation, equipment operation, and material circulation in the production process, which is an important data basis for enterprises to realize production compliance supervision, safety risk early warning, and production efficiency optimization. The steps of the preprocessing operation are as follows: first, noise removal, through mean filtering, median filtering, Gaussian filtering, and other algorithms, to eliminate the speckles and noise points in the image caused by electronic component interference, light fluctuation, dust particles, and the like, to make the image overall smoother; second, image enhancement, through histogram equalization, contrast stretching, and other methods to improve the brightness, contrast, and color saturation of the image, to make the difference between the target area and the background in the image more obvious, facilitating target identification; third, geometric correction, for image geometric distortion caused by installation angle deviation of the monitoring device, lens distortion, or vibration, through coordinate transformation, perspective correction, and other means to restore the image to the correct geometric shape, to ensure that the size, position, and other information of the object in the image are accurate; and finally, image cropping and scaling, according to actual monitoring requirements, to crop the irrelevant background area in the image and retain the effective part containing key production information, and at the same time, according to the requirements of the subsequent processing algorithm, to scale the image to an appropriate size, to reduce the data processing amount and improve the speed of subsequent analysis.

[0023] In an implementation, after inputting the optimization parameters into the preset chaotic mapping, an iteration process is started, and the number of iterations needs to be determined according to the pixel size of the target sub-original image. The number of iterations is set to (200+N) times, where N is the total number of image pixels. First, 200 initial iterations are performed and the results of this part are discarded to eliminate the influence of initial value fluctuation on the randomness of the sequence. Then, the iteration continues until a chaotic sequence set adapted to the pixel number of the target sub-original image is generated. During the generation process, the randomness and complexity of the chaotic sequence need to be ensured, and finally a chaotic sequence set containing multiple independent chaotic sequences is obtained, providing a basis for subsequent key generation. In an implementation, the preset chaotic mapping and the preset combination rule are determined by technical personnel.

[0024] In one implementation, after the target sub-original image is encrypted to obtain a sub-encrypted image, the sub-encrypted image is stored in different physical locations according to a preset storage rule, and the sub-encrypted image is stored in different physical locations according to the sensitivity of the production monitoring image data of the industrial and trade enterprise. These locations need to meet the geographical dispersion, such as local servers of enterprises, off-site backup centers, and third-party secure storage nodes, and the diversity of storage media, such as hard disk arrays, tape libraries, and cloud storage nodes, to avoid overall data security problems caused by single location failure or leakage. During storage, a unique identifier and storage location mapping relationship needs to be established for each sub-encrypted image data block, the integrity check information of the data block is recorded, and the access permission of the storage location is set. Only authorized personnel can obtain the data through verification to ensure the security and traceability of the sub-encrypted image during multi-physical location storage.

[0025] In one implementation, after the sub-encrypted image storage is completed, the transmission process is started. First, it is clear that the preset transmission path needs to meet the security and stability requirements of the industrial and trade enterprise data transmission, and the transmission channel is constructed by preferentially selecting an encrypted transmission protocol to avoid data interception or tampering during transmission. Then, the information to be transmitted is sorted, including the storage path of the sub-encrypted image, which needs to include the access address and data block identifier of each physical location, the scrambling control key and the diffusion control key. During transmission, the transmission state needs to be monitored in real time, and the data fragmentation transmission and retransmission mechanism is set. If a certain fragment transmission fails, retransmission can be triggered based on the integrity check information to ensure the integrity of data transmission. After the cloud receiver receives the transmission information, identity verification and information integrity check are needed. After the check is passed, the cloud stores the sub-encrypted image storage path, the scrambling control key and the diffusion control key, and establishes an associated index, so as to quickly locate the sub-encrypted image storage location and call the corresponding key during subsequent authorized decryption. At the same time, transmission logs are recorded, including transmission time, transmission node, operator and other information, realizing the auditability of the transmission process, meeting the security traceability requirements of medical image encryption transmission in the document, and adapting to the security management needs of the production monitoring image data of the industrial and trade enterprise.

[0026] In one embodiment, the original image is preliminarily divided to obtain a set of preliminary sub-original images; The information redundancy of the preliminary sub-images in the set of preliminary sub-images is calculated to obtain a sub-image information redundancy; The quantum correlation degree between each preliminary sub-image in the set of preliminary sub-images is calculated to obtain a sub-image correlation degree; The initial division parameters are adjusted according to the sub-image information redundancy and the sub-image correlation degree to obtain target division parameters; The original image is divided according to the target division parameters to obtain a set of sub-original images.

[0027] In an implementation, for each preliminary sub-image in the preliminary sub-image set, the pixel value distribution features such as mean, variance, entropy, etc. of the gray value are extracted, and a redundancy calculation method in information theory is used. The redundancy calculation method can be a statistical analysis based on the correlation between pixels, and the difference between the joint probability distribution and the edge probability distribution of adjacent pixels is calculated to quantify the degree of information repetition within each sub-image. Specifically, the spatial redundancy and visual redundancy of the sub-image can be calculated. The spatial redundancy calculation method can be the proportion of the gray value difference between adjacent pixels that is less than a threshold value, and the visual redundancy calculation method can be to analyze the proportion of pixels that contribute less to visual perception based on the visual characteristics of the human eye. These indicators are weighted and fused to obtain the information redundancy value of each preliminary sub-image. The higher the value, the more repeated the information within the sub-image, which provides a basis for subsequent parameter adjustment.

[0028] In an implementation, any two preliminary sub-images in the preliminary sub-image set are selected as the calculation object, and features such as color histogram, texture feature matrix, and key pixel coordinate set are extracted. By referring to the calculation idea of correlation degree in quantum information science, the feature vectors are mapped to quantum state vectors, and the internal correlation degree between them is quantified by calculating the entanglement entropy or mutual information of the two quantum state vectors. For multiple sub-images, all two-by-two combinations need to be calculated to obtain a correlation degree matrix. The higher the element value in the matrix, the stronger the correlation between the corresponding two sub-images in terms of content features, which reflects whether the preliminary division has excessively divided the regions that should be associated.

[0029] In an implementation, information redundancy threshold and correlation threshold are set. First, preliminary sub-images with information redundancy exceeding the threshold are screened out. For such sub-images, the division granularity of the region where they are located is increased to reduce internal redundancy. At the same time, sub-image pairs with correlation exceeding the threshold are screened out. Their positional relationship in the original image is analyzed. If they belong to adjacent regions, the division granularity of the corresponding region is reduced or the segmentation boundary is adjusted so that the content with strong correlation is as much as possible in the same sub-image. Through multiple rounds of iteration adjustment of initial division parameters such as grid row and column numbers and segmentation threshold, the information redundancy and correlation are recalculated after each round of adjustment until the information redundancy of most sub-images is below the threshold and the number of high-correlation sub-image pairs is below the threshold. At this time, the division parameters are the target division parameters.

[0030] In one embodiment, the original image is subjected to resolution and information entropy extraction to obtain an image feature vector; The original image is subjected to quantumization simulation processing to obtain a quantum feature vector; The image feature vector and the quantum feature vector are input into a preset neural network model of a fusion quantum calculation module to obtain initial division parameters; The original image is preliminarily divided according to the initial division parameters to obtain a preliminary sub-image set.

[0031] In an implementation, first, the resolution parameter of the original image is determined, the width and height pixel values are obtained by reading the image metadata, and the total pixel number is calculated to quantify the size characteristics of the image; then, an information entropy calculation method is used, which can be a Shannon entropy formula based on the probability distribution of gray levels, to calculate the probability of each gray level appearing in the image, and the information entropy is calculated by the accumulation of the product of the probability value and the logarithmic probability, to reflect the richness of the image information. The resolution parameter and the information entropy value are standardized, and combined in a predetermined order to form a one-dimensional array, which is the image feature vector that can represent the basic characteristics of the original image, providing basic data for subsequent model input.

[0032] In an implementation, first, the pixel values of the original image are normalized and mapped to the value range of the quantum state space; then, based on the principle of quantum computing, the two-dimensional pixel matrix of the image is converted into a mathematical representation of the quantum superposition state, for example, by encoding each pixel value as the probability amplitude of a quantum bit, to construct a quantum state vector containing the spatial distribution information of the image. Further, the quantum state vector is measured and simulated, the density matrix, entanglement entropy and other quantum characteristic parameters of the quantum state are calculated, and the key indicators reflecting the quantumization characteristics of the image are extracted. These indicators are arranged in a specific dimension to form a quantum feature vector with quantum characteristic representation, which can capture the deep correlation information of the image that traditional features cannot express.

[0033] In an implementation, the preset neural network model adopts a fusion architecture, the input layer of which is designed as a double-channel structure to receive the image feature vector and the quantum feature vector respectively, and the two are fused into a high-dimensional joint feature vector through a feature splicing layer; the middle layer of the model includes a quantum computing module, which performs quantum state transformation and entanglement processing on the joint features through quantum gate operations, enhances the non-linear association mining capability between features, and then converts the quantum features into classical feature vectors through measurement operations; the output layer of the model adopts a fully connected structure, and according to the typical segmentation requirements of the production monitoring images of industrial and trade enterprises, outputs an initial segmentation parameter set containing the number of grid rows, the number of grid columns, the key region segmentation threshold and other parameters. The parameter values are mapped to a reasonable value range through an activation function, to ensure that the basic constraints of image segmentation are met.

[0034] In one embodiment, a candidate parameter set and an adaptability function of a chaotic optimization algorithm are constructed; the candidate parameters in the candidate parameter set are generated by a random function within a preset value range; the adaptability function is established according to the information entropy and the pixel correlation index of the image; The target candidate parameter is input into the preset chaotic mapping to obtain a chaotic sequence; the target candidate parameter is any one of the candidate parameters in the candidate parameter set; The historical original image is encrypted according to a historical scrambling control key, a historical diffusion control key and a chaotic sequence to obtain a historical encrypted image. The fitness value of the historical encrypted image is calculated according to an adaptive function to obtain a candidate parameter fitness value. The candidate parameter with the maximum candidate parameter fitness value in the candidate parameter set is selected as the optimized parameter. In an implementation manner, the preset value range and the preset chaotic mapping are set by a technician.

[0035] In an implementation manner, when the candidate parameter set is constructed, the parameter types of the preset chaotic mapping are first determined, including initial state variables A1, A2, A3, A4, A5, A6, A7, coupling parameters W, stable parameters S, T, P and control parameters R, E, I, Q1, Q2, G, and the preset value ranges of the parameters are set, for example, the value range of the control parameter R can be set as [0.5, 2.0], and the value range of the stable parameter S can be set as [8, 12]. Then, a uniform distribution random function is called to generate multiple groups of independent parameter combinations in the preset value ranges of the parameters, each group of parameter combinations contains the required parameters of the preset chaotic mapping, and all the parameter combinations together constitute the candidate parameter set. When the adaptive function is constructed, the maximum image information entropy and the minimum pixel correlation are taken as the targets, and a weighted combination method is used to establish the function: first, an information entropy function h(p) is defined, which is used to calculate the Shannon entropy of the probability distribution of the image gray value, and the ideal value is close to 8; a pixel correlation function Co(p) is used to calculate the correlation coefficients of adjacent pixels in the horizontal, vertical and diagonal directions, and the ideal value is close to 0; then, weight coefficients are set, for example, W1=0.1 and W2=0.9; and finally, the adaptive function f(p)=W1×Co(p)+W2×(8−h(p)) is constructed, and the smaller the function value is, the better the encryption effect of the corresponding parameter is. In an implementation manner, the iteration formula of the preset chaotic mapping containing seven state variables is A1′=S×(A2−A1)+A4+R×A6, A2′=P×A1−A2−A1×A3+A5, A3'= -TxA3 + 1xA2, A4'= ExA4 - 1xA3, A5'= IxA2 + A6, A6'= Q1xA1 + Q2xA2, A7'= GxA7 + NxA4; wherein, S is a stable parameter, controlling the diffusion coefficient of state variable A1, and the default value is 10; P is a stable parameter, controlling the linear term coefficient of state variable A2, and the default value is 28; T is a stable parameter, controlling the attenuation coefficient of state variable A3; R is a control parameter, adjusting the contribution weight of A6 to A1', and the default value is 1; E is a control parameter, controlling the self-iteration attenuation coefficient of A4, and the default value is 2; I is a control parameter, adjusting the contribution weight of A2 to A5', and the default value is 9.9; Q1 is a control parameter, adjusting the contribution weight of A1 to A6', and the default value is 1; Q2 is a control parameter, adjusting the contribution weight of A2 to A6', and the default value is 2; G is a control parameter, controlling the self-iteration coefficient of A7, and the default value is 1; and N is a coupling parameter, adjusting the coupling influence of A4 on A7'. Selecting any one group of target candidate parameters in the candidate parameter set, the target candidate parameters including initial state variables, coupling parameters, stable parameters and control parameters required by the preset chaotic mapping, setting the iteration times in the preset chaotic mapping, and adopting the strategy of iterating (200+N) times and discarding the results of the first 200 times, N being the total number of pixels of the historical original image, first performing 200 initial iterations to eliminate the influence of initial value fluctuation on sequence randomness, and then continuing to iterate N times to generate a chaotic sequence matching the pixel number of the historical original image. During the iteration process, the values of the seven state variables output by each iteration are recorded in real time, and these values are arranged into seven independent one-dimensional numerical sequences in the order of iteration, each sequence being a chaotic sequence in the chaotic sequence set, and finally forming a complete chaotic sequence set.

[0036] In an implementation manner, the historical scrambling control key and the historical diffusion control key are generated by the technicians in combination with the resolution characteristics and pixel distribution rules of the common original images of the industrial and trade enterprises.

[0037] In an embodiment, the target sub-original image is subjected to a gray value binary conversion to obtain a one-dimensional sequence. The one-dimensional sequence is subjected to scrambling to obtain a scrambled one-dimensional sequence; wherein, is the binary value of the element in the one-dimensional sequence, is the next state obtained by the transformation of the element in the one-dimensional sequence, is the binary value of the previous element of the element in the one-dimensional sequence, is the binary value of the next element of the element in the one-dimensional sequence, is the binary value of the element in the one-dimensional sequence, ​​​​​To scramble the control key The binary value of each element. It is the XOR operator; it converts a scrambled one-dimensional sequence back to a grayscale array to obtain a scrambled sub-image. Sub-encrypted images are obtained by forward and reverse diffusion of the scrambled image based on the diffusion control key.

[0038] One implementation involves first converting the original target sub-image to grayscale to obtain a one-dimensional sequence, then scrambling the one-dimensional sequence using a specific XOR operation combined with a scrambling control key, subsequently converting the scrambled one-dimensional sequence back to a grayscale array to obtain a scrambled sub-image, and finally performing forward and reverse diffusion on the scrambled image according to the diffusion control key to obtain a sub-encrypted image. This series of processes fully utilizes the characteristics of XOR operations and the roles of scrambling and diffusion control keys, significantly improving the security and complexity of image encryption, making the encrypted image more difficult to crack illegally, and effectively protecting important information such as security risks involved in images related to industrial and commercial enterprises.

[0039] Based on the same inventive concept, this invention also provides an intelligent management system for safety risks in industrial and commercial enterprises. See also... Figure 2 , Figure 2 A framework diagram of an intelligent management system for safety risks in industrial and commercial enterprises, provided as an embodiment of the present invention, includes: The raw image acquisition module is used to acquire raw images of industrial and trading enterprises; the raw images are pre-processed production monitoring images of the enterprises. The segmentation module is used to segment the original image to obtain a sub-set of original images; The parameter update module is used to update the parameters of the preset chaotic mapping based on historical data to obtain optimized parameters; The chaotic sequence generation module is used to input optimization parameters into a preset chaotic mapping and iterate to obtain a chaotic sequence set; The key generation module is used to obtain scrambling control keys and diffusion control keys from chaotic sequences in the chaotic sequence set according to preset combination rules; The sub-encrypted image generation module is used to convert and encrypt the target sub-original image according to the scrambling control key and the diffusion control key to obtain the sub-encrypted image; the target sub-original image is any one of the sub-original images in the set of sub-original images. The storage module is used to store the sub-encrypted images in different physical locations according to preset storage rules; The transmission module is used to transmit the sub-encrypted image storage path, scrambling control key, and diffusion control key to the cloud via a preset transmission path.

[0040] Based on the embodiment of the application, the original image of the industrial and trade enterprise is first obtained, and a set of sub-original images is segmented; then the preset chaotic mapping parameters are updated according to the historical data, and the optimized parameters are input into the mapping iteration to generate a set of chaotic sequences; then the scrambling and diffusion control keys are extracted from the sequences according to the preset rules, and the target sub-original image is converted and encrypted by using the two kinds of keys to obtain a sub-encrypted image; then the sub-encrypted image is stored in different physical locations according to the preset rules, and finally the storage path and the two kinds of keys are transmitted to the cloud through the preset transmission path. According to the historical data, the parameter optimization improves the adaptability and security of the key, the chaotic encryption enhances the complexity and anti-cracking ability, and the sub-image dispersed storage reduces the risk of leakage, which comprehensively ensures the safety of the whole process of enterprise production monitoring image data.

[0041] In one embodiment, the segmentation module comprises: A preliminary sub-original image generation module is configured to preliminarily divide the original image to obtain a set of preliminary sub-original images. A redundancy calculation module is configured to calculate the information redundancy of the preliminary sub-images in the set of preliminary sub-images to obtain the sub-image information redundancy. A correlation calculation module is configured to calculate the quantum correlation between each preliminary sub-image in the set of preliminary sub-images to obtain the sub-image correlation. A target division parameter generation module is configured to adjust the initial division parameters according to the sub-image information redundancy and the sub-image correlation to obtain the target division parameters. A division module is configured to divide the original image according to the target division parameters to obtain the set of sub-original images.

[0042] In one embodiment, the preliminary sub-original image generation module comprises: An image feature extraction module is configured to extract the resolution and information entropy of the original image to obtain an image feature vector. A quantum feature vector generation module is configured to perform quantum simulation processing on the original image to obtain a quantum feature vector. An initial division parameter generation module is configured to input the image feature vector and the quantum feature vector into a preset neural network model of the fusion quantum calculation module to obtain the initial division parameters. A preliminary division module is configured to preliminarily divide the original image according to the initial division parameters to obtain the set of preliminary sub-images.

[0043] In one embodiment, the parameter update module comprises: A parameter function construction module is configured to construct a candidate parameter set and an adaptability function of a chaotic optimization algorithm; the candidate parameters in the candidate parameter set are generated by a random function within a preset value range; and the adaptability function is established according to the information entropy and the pixel correlation index of the image. The chaotic sequence generation module is used to input the target candidate parameter into a preset chaotic mapping to obtain a chaotic sequence; the target candidate parameter is any one of the candidate parameters in the candidate parameter set; The historical encrypted image generation module is used to encrypt the original historical image based on the historical scrambling control key, the historical diffusion control key, and the chaotic sequence to obtain the historical encrypted image; The fitness value calculation module is used to calculate the fitness value of the candidate parameters based on the fitness function for historical encrypted images. The optimization parameter generation module is used to select the candidate parameter with the largest fitness value corresponding to the candidate parameter in the candidate parameter set as the optimization parameter.

[0044] In one embodiment, the sub-encrypted image generation module includes: The one-dimensional sequence generation module is used to convert the grayscale values ​​of the original target image to binary to obtain a one-dimensional sequence. The scrambling operation module is used to first pass through Then through Scrambling a one-dimensional sequence yields a scrambled one-dimensional sequence; where, For the first one-dimensional sequence The next state obtained after transforming each element For the first one-dimensional sequence The binary value of the element preceding the element. For the first one-dimensional sequence The binary value of the next element after the first element. For the first one-dimensional sequence The binary value of each element. To scramble the control key The binary value of each element. It is the XOR operator; the scrambled sub-image generation module is used to convert a scrambled one-dimensional sequence back to a grayscale array to obtain a scrambled sub-image. The sub-encrypted image generation module is used to perform forward and reverse diffusion on the scrambled image according to the diffusion control key to obtain the sub-encrypted image.

[0045] The foregoing detailed one embodiment of the present invention, but this is merely a preferred embodiment and should not be construed as limiting the scope of the invention. All equivalent variations and modifications made within the scope of the present invention should still fall within the patent coverage of the present invention.

Claims

1. A method for intelligent management of safety risks in industrial enterprises, characterized in that, The method comprises: acquiring an original image of an industrial and trade enterprise; the original image is a pre-processed enterprise production monitoring image; segmenting the original image to obtain a set of sub-original images; updating parameters of a preset chaotic mapping according to historical data to obtain optimized parameters; inputting the optimized parameters into the preset chaotic mapping for iteration to obtain a set of chaotic sequences; obtaining a scrambling control key and a diffusion control key according to a preset combination rule for chaotic sequences in the set of chaotic sequences; performing data conversion and encryption on a target sub-original image according to the scrambling control key and the diffusion control key to obtain a sub-encrypted image; the target sub-original image is any one of the set of sub-original images; storing the sub-encrypted image in different physical locations according to a preset storage rule; transmitting the sub-encrypted image storage path, the scrambling control key and the diffusion control key to the cloud through a preset transmission path.

2. The intelligent safety risk management method for industrial enterprises according to claim 1, characterized in that, Segmenting the original image to obtain a set of sub-original images comprises: preliminarily dividing the original image to obtain a set of preliminary sub-original images; calculating information redundancy of the preliminary sub-images in the set of preliminary sub-images to obtain sub-image information redundancy; calculating quantum correlation degree between each preliminary sub-image in the set of preliminary sub-images to obtain sub-image correlation degree; adjusting initial division parameters according to the sub-image information redundancy and the sub-image correlation degree to obtain target division parameters; dividing the original image according to the target division parameters to obtain a set of sub-original images.

3. The intelligent management method of safety risks of industrial enterprises according to claim 2, characterized in that, Preliminarily dividing the original image to obtain a set of preliminary sub-original images comprises: extracting image feature vectors from the original image according to resolution and information entropy; performing quantum simulation processing on the original image to obtain quantum feature vectors; inputting the image feature vectors and the quantum feature vectors into a preset neural network model of a fusion quantum computing module to obtain initial division parameters; preliminarily dividing the original image according to the initial division parameters to obtain a set of preliminary sub-images.

4. The intelligent safety risk management method for industrial enterprises according to claim 1, characterized in that, Updating parameters of a preset chaotic mapping according to historical data to obtain optimized parameters comprises: constructing a candidate parameter set and an adaptability function of the chaotic optimization algorithm; the candidate parameters in the candidate parameter set are generated by a random function within a preset value range; the adaptability function is established according to the information entropy and pixel correlation index of the image; inputting a target candidate parameter into a preset chaotic mapping to obtain a chaotic sequence; the target candidate parameter is any one of the candidate parameters in the candidate parameter set; encrypting historical original images according to historical scrambling control keys, historical diffusion control keys and the chaotic sequence to obtain historical encrypted images; calculating the fitness value of the historical encrypted images according to the adaptability function to obtain candidate parameter fitness values; selecting a candidate parameter in the candidate parameter set as an optimized parameter, the candidate parameter corresponding to the maximum candidate parameter fitness value.

5. The intelligent safety risk management method for industrial enterprises according to claim 1, characterized in that, Performing data conversion and encryption on a target sub-original image according to the scrambling control key and the diffusion control key to obtain a sub-encrypted image comprises: The original image of the target sub-image is converted from grayscale to binary to obtain a one-dimensional sequence; firstly, through... Then through The one-dimensional sequence is scrambled to obtain a scrambled one-dimensional sequence; wherein, For the first one-dimensional sequence The next state obtained after transforming each element For the first one-dimensional sequence The binary value of the element preceding the element. For the first one-dimensional sequence The binary value of the next element after the first element. For the first one-dimensional sequence The binary value of each element. The scrambling control key is the first one. The binary value of each element. It is the XOR operator; the scrambled one-dimensional sequence is converted back to a grayscale array to obtain the scrambled sub-image; performing forward and reverse diffusion on the scrambled image according to the diffusion control key to obtain a sub-encrypted image.

6. A security risk intelligent management system for industrial and trade enterprises, characterized in that, The system comprises: An original image acquisition module for acquiring an original image of an industrial and trade enterprise; the original image is a pre-processed enterprise production monitoring image; A segmentation module for segmenting the original image to obtain a set of sub-original images; A parameter updating module for updating parameters of a preset chaotic mapping according to historical data to obtain optimized parameters; A chaotic sequence generation module for inputting the optimized parameters into the preset chaotic mapping for iteration to obtain a set of chaotic sequences; A key generation module for obtaining a scrambling control key and a diffusion control key from chaotic sequences in the set of chaotic sequences according to a preset combination rule; A sub-encrypted image generation module for data conversion and encryption of a target sub-original image according to the scrambling control key and the diffusion control key to obtain a sub-encrypted image; the target sub-original image is any one of the set of sub-original images; A storage module for storing the sub-encrypted image in different physical locations according to a preset storage rule; A transmission module for transmitting the sub-encrypted image storage path, the scrambling control key and the diffusion control key to the cloud through a preset transmission path.

7. The intelligent safety risk management system for industrial enterprises according to claim 6, characterized in that, The segmentation module comprises: A preliminary sub-original image generation module for preliminarily dividing the original image to obtain a set of preliminary sub-original images; A redundancy calculation module for calculating information redundancy of preliminary sub-images in the set of preliminary sub-images to obtain sub-image information redundancy; A correlation degree calculation module for calculating quantum correlation degrees between each preliminary sub-image in the set of preliminary sub-images to obtain sub-image correlation degrees; A target division parameter generation module for adjusting initial division parameters according to the sub-image information redundancy and the sub-image correlation degrees to obtain target division parameters; A division module for dividing the original image according to the target division parameters to obtain the set of sub-original images.

8. The intelligent safety risk management method for industrial enterprises according to claim 7, characterized in that, The preliminary sub-original image generation module comprises: An image feature extraction module for extracting resolution and information entropy of the original image to obtain an image feature vector; A quantum feature vector generation module for quantum analog processing of the original image to obtain a quantum feature vector; An initial division parameter generation module for inputting the image feature vector and the quantum feature vector into a preset neural network model of a fusion quantum calculation module to obtain initial division parameters; A preliminary division module for preliminarily dividing the original image according to the initial division parameters to obtain a set of preliminary sub-images.

9. The intelligent safety risk management method for industrial enterprises according to claim 6, characterized in that, The parameter updating module comprises: A parameter function construction module for constructing a candidate parameter set and an adaptability function of the chaotic optimization algorithm; the candidate parameters in the candidate parameter set are generated by a random function within a preset value range; the adaptability function is established according to information entropy and pixel correlation indexes of an image; A chaotic sequence generation module for inputting a target candidate parameter into a preset chaotic mapping to obtain a chaotic sequence; the target candidate parameter is any one of the candidate parameters in the candidate parameter set; a history encryption image generation module configured to encrypt a history original image according to a history scrambling control key, a history diffusion control key and the chaotic sequence to obtain a history encryption image; a fitness value calculation module configured to calculate a fitness value of the history encryption image according to the adaptive function to obtain a candidate parameter fitness value; an optimized parameter generation module configured to select a candidate parameter with the maximum candidate parameter fitness value in the candidate parameter set as an optimized parameter.

10. The intelligent safety risk management method for industrial enterprises according to claim 6, characterized in that, The sub-encryption image generation module comprises: A one-dimensional sequence generation module is used to perform grayscale-to-binary conversion on the original target sub-image to obtain a one-dimensional sequence; a scrambling operation module is used to first... Then through The one-dimensional sequence is scrambled to obtain a scrambled one-dimensional sequence; wherein, For the first one-dimensional sequence The next state obtained after transforming each element For the first one-dimensional sequence The binary value of the element preceding the element. Let be the binary value of the element following the i-th element in the one-dimensional sequence. For the first one-dimensional sequence The binary value of each element. The scrambling control key is the first one. The binary value of each element. It is the XOR operator; a scrambled sub-image generation module configured to convert the scrambled one-dimensional sequence back to a gray value array to obtain a scrambled sub-image; a sub-encryption image generation module configured to perform forward and reverse diffusion on the scrambled image according to the diffusion control key to obtain a sub-encryption image.

Images