Distributed storage method for digital image data

Abstract

The invention discloses a distributed storage method of digital image data, and belongs to the technical field of data storage, a cluster server is adopted to construct a distributed storage system comprising a main storage node and an edge storage node, the main storage node adopts a fragmentation encryption mode to encrypt the digital image data, and the edge storage node adopts a fragmentation encryption mode to encrypt the digital image data. The encrypted digital image data and a plurality of fragmented encryption keys are obtained, the encrypted digital image data are randomly transmitted to two first edge storage nodes through a main storage node, and each fragmented encryption key is transmitted to a different second edge storage node; decentralization and double backup storage can be effectively realized, and the secure storage of digital image data is greatly improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of data storage, and in particular relates to a distributed storage method for digital image data. Background Art

[0002] Digital image data refers to image information that is captured, processed, stored, and transmitted in a digital way. These data can be in the form of photos, videos, medical images, satellite images, etc. They usually exist in the form of digital files and can be operated and processed by computers or other electronic devices. With the continuous development of digital imaging technology, the amount of digital image data has shown explosive growth. Although the traditional centralized storage method can centrally manage data, it has the problem of poor security. Summary of the invention

[0003] The present invention provides a distributed storage method for digital image data, which is used to solve the problem of low storage security performance in the prior art.

[0004] A distributed storage method for digital image data, comprising:

[0005] A distributed storage system is constructed using a cluster server; wherein the distributed storage system includes a main storage node and an edge storage node;

[0006] Obtaining a data image data storage request through a primary storage node, and verifying the data image data storage request to obtain a request verification result; wherein the request verification result includes verification passed or verification failed; the data image data storage request includes patient information and digital image data;

[0007] The digital image data is encrypted by using a fragmented encryption method through a main storage node to obtain the encrypted digital image data and a plurality of fragmented encryption keys;

[0008] The encrypted digital image data is randomly transmitted to two first edge storage nodes through the main storage node, and each fragmented encryption key is transmitted to a different second edge storage node;

[0009] The data storage address is returned through the first edge storage node and the key storage address is returned through the second edge storage node, and the patient information, the data storage address and the key storage address are associated and stored through the main storage node.

[0010] Furthermore, a distributed storage system is constructed using a cluster server, including:

[0011] Based on the cluster servers, determine the sum of the communication distances between each server and all other servers;

[0012] According to the sum of the communication distances between each server and all other servers, the server with the smallest sum of communication distances is determined as the main storage node, and the other servers are determined as edge storage nodes.

[0013] Further, obtaining a data image data storage request through the main storage node, and verifying the data image data storage request to obtain a request verification result, including:

[0014] Obtaining a data image data storage request through a main storage node, and obtaining patient information to be stored and digital image data to be stored;

[0015] Performing online verification on the name, gender and ID card number in the stored patient information to obtain a basic verification result; wherein the basic verification result includes verification success or verification failure;

[0016] When the basic verification result is verification failure, determining the request verification result is verification failure;

[0017] When the basic verification result is a successful verification, a missing verification is performed on the inspection information and the inspection sequence information to obtain a missing verification result; wherein the missing verification result includes a successful verification or a failed verification;

[0018] When the missing verification result is a successful verification, the request verification result is determined to be a passed verification, otherwise the request verification result is determined to be a failed verification.

[0019] Furthermore, the digital image data is encrypted by the main storage node using a fragmented encryption method to obtain the encrypted digital image data and a plurality of fragmented encryption keys, including:

[0020] The digital image data is encrypted using a symmetric encryption algorithm to obtain the encrypted digital image data and the corresponding symmetric encryption key;

[0021] Based on the symmetric encryption key, a fragmentation processing polynomial is generated, and a plurality of fragmentation encryption keys are obtained according to the fragmentation processing polynomial.

[0022] Furthermore, based on the symmetric encryption key, a fragmentation processing polynomial is generated as follows:

[0023] The symmetric encryption key is converted into a decimal number, and a fragmentation processing polynomial is constructed according to the converted decimal number:

[0024] f(x)=s+a1x+a2x 2 +…+a k x k

[0025] Among them, f(x) represents the dependent variable, x represents the independent variable, k represents the order of the polynomial, a1, a2,…, a k They represent the 1st, 2nd, …, kth different fragmentation processing polynomial coefficients respectively, and s represents a decimal number.

[0026] Further, obtaining a plurality of fragmented encryption keys according to the fragmentation processing polynomial includes:

[0027] The converted decimal number is divided into at least 2k key fragments, and each key fragment is brought into the fragmentation processing polynomial to obtain the target dependent variable corresponding to each key fragment;

[0028] For any key fragment, the key fragment and the target dependent variable are combined into a fragmented encryption key to obtain multiple fragmented encryption keys.

[0029] Furthermore, the encrypted digital image data is randomly transmitted to two first edge storage nodes through the main storage node, and each fragmented encryption key is transmitted to a different second edge storage node, including:

[0030] Randomly determine two first edge storage nodes through the main storage node, randomly transmit the encrypted digital image data to the two first edge storage nodes, and set the remaining edge storage nodes as second edge storage nodes; wherein the number of the second edge storage nodes is greater than 2k;

[0031] 2k second edge storage nodes are randomly determined, and each fragmented encryption key is transmitted to a different second edge storage node.

[0032] Furthermore, it also includes:

[0033] Obtaining a data query request input by a user; wherein the data query request includes at least one item of patient information;

[0034] According to the data query request, searching the corresponding target patient information, target data storage address and target key storage address through the main storage node;

[0035] According to the target data storage address and the target key storage address, the encrypted digital image data and the fragmented encryption key are obtained, and encrypted and transmitted through the public key of the user device.

[0036] Furthermore, while obtaining the data query request input by the user, it also includes:

[0037] Collect network traffic features, and use a pre-deployed deep learning model to detect the network traffic features to obtain network security detection results; wherein the network security detection results include network security or network intrusion;

[0038] When the network security detection result shows that the network has been invaded, a network security protection operation is performed to ensure the storage security of the digital image data.

[0039] Furthermore, the pre-deployment method of the deep learning model includes:

[0040] A population of algorithms is initialized using a uniform initialization strategy, where each individual in the population consists of all or part of the hyperparameters of the deep learning model;

[0041] Perform exploration phase spatial search, development phase spatial search, encirclement phase spatial search and escape phase spatial search on the population in sequence to obtain the population after multi-stage spatial search;

[0042] Determine whether the population after the multi-stage spatial search meets the iteration end condition. If so, select the best individual from the population after the multi-stage spatial search and preset the deep learning model with the best individual. Otherwise, return to the multi-stage search step.

[0043] The present invention provides a distributed storage method for digital image data. A distributed storage system including a main storage node and an edge storage node is constructed by using a cluster server. The digital image data is encrypted by the main storage node using a fragmented encryption method to obtain the encrypted digital image data and multiple fragmented encryption keys. The encrypted digital image data is randomly transmitted to two first edge storage nodes through the main storage node, and each fragmented encryption key is transmitted to a different second edge storage node. Decentralization and dual backup storage can be effectively realized, which greatly improves the secure storage of digital image data. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0045] Figure 1 The present invention provides a flowchart of a distributed storage method for digital image data.

[0046] Figure 2 A flowchart of a method for pre-deploying a deep learning model provided in an embodiment of the present invention.

[0047] The above drawings have shown clear embodiments of the present invention, which will be described in more detail below. These drawings and text descriptions are not intended to limit the scope of the present invention in any way, but to illustrate the concept of the present invention for those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION

[0048] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present invention. Instead, they are merely examples of devices and methods consistent with some aspects of the present invention as detailed in the appended claims.

[0049] The embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0050] like Figure 1 As shown, an embodiment of the present invention provides a distributed storage method for digital image data, comprising:

[0051] S101, using a cluster server to build a distributed storage system; wherein the distributed storage system includes a main storage node and an edge storage node;

[0052] Building a distributed storage system through cluster servers can not only effectively improve data storage capabilities, but also achieve decentralized processing and improve data security storage capabilities.

[0053] S102, obtaining a data image data storage request through a primary storage node, and verifying the data image data storage request to obtain a request verification result; wherein the request verification result includes verification passed or verification failed; the data image data storage request includes patient information and digital image data;

[0054] In order to ensure that the stored data is valid, it is necessary to first verify the data image data storage request to avoid data recording errors or omissions.

[0055] S103, encrypting the digital image data by using a fragmented encryption method through the main storage node to obtain the encrypted digital image data and a plurality of fragmented encryption keys;

[0056] By obtaining encrypted digital image data and multiple fragmented encryption keys, the key can be made more inaccessible. Even if an illegal intruder obtains part of the data or the key, he cannot decrypt it, thus effectively ensuring the security of the algorithm.

[0057] S104, randomly transmitting the encrypted digital image data to two first edge storage nodes through the main storage node, and transmitting each fragmented encryption key to a different second edge storage node;

[0058] For the digital imaging data corresponding to any patient, backup processing is performed, which can effectively ensure the security of the data and resist network attacks and physical attacks. Each fragmented encryption key is transmitted to a different second edge storage node, which can increase the difficulty of obtaining the fragmented encryption key. During the transmission process, the public key of the second edge storage node is used for encrypted transmission to prevent the fragmented encryption key from being obtained. The complexity of the system increases exponentially, which can effectively improve data security.

[0059] S105. Return the data storage address through the first edge storage node and return the key storage address through the second edge storage node, and associate and store the patient information, the data storage address, and the key storage address through the main storage node.

[0060] When associated storage is performed, the data can be encrypted, and when the user accesses it normally, the data is decrypted and the related data is obtained, thereby realizing a double encryption mechanism.

[0061] In an embodiment of the present invention, a distributed storage system is constructed using a cluster server, including:

[0062] Based on the cluster servers, determine the sum of the communication distances between each server and all other servers;

[0063] According to the sum of the communication distances between each server and all other servers, the server with the smallest sum of communication distances is determined as the main storage node, and the other servers are determined as edge storage nodes.

[0064] Since the main storage node needs to communicate with other edge storage nodes, the server with the smallest total communication distance is determined as the main storage node, thereby improving data storage and reading efficiency.

[0065] In an embodiment of the present invention, a data image data storage request is obtained through a primary storage node, and the data image data storage request is verified to obtain a request verification result, including:

[0066] Obtaining a data image data storage request through a main storage node, and obtaining patient information to be stored and digital image data to be stored;

[0067] Performing online verification on the name, gender and ID card number in the stored patient information to obtain a basic verification result; wherein the basic verification result includes verification success or verification failure;

[0068] When the basic verification result is verification failure, determining the request verification result is verification failure;

[0069] When the basic verification result is a successful verification, a missing verification is performed on the inspection information and the inspection sequence information to obtain a missing verification result; wherein the missing verification result includes a successful verification or a failed verification;

[0070] When the missing verification result is a successful verification, the request verification result is determined to be a passed verification, otherwise the request verification result is determined to be a failed verification.

[0071] In an embodiment of the present invention, the digital image data is encrypted by the main storage node using a fragmented encryption method to obtain the encrypted digital image data and a plurality of fragmented encryption keys, including:

[0072] The digital image data is encrypted using a symmetric encryption algorithm to obtain the encrypted digital image data and the corresponding symmetric encryption key;

[0073] Based on the symmetric encryption key, a fragmentation processing polynomial is generated, and a plurality of fragmentation encryption keys are obtained according to the fragmentation processing polynomial.

[0074] In the embodiment of the present invention, based on the symmetric encryption key, the fragmentation processing polynomial is generated as follows:

[0075] The symmetric encryption key is converted into a decimal number (for example, when the Advanced Encryption Standard algorithm is used, 11 rounds of keys are required, and the 11 rounds of keys can be arranged in sequence and then converted into a decimal number), and a fragmentation processing polynomial is constructed according to the converted decimal number:

[0076] f(x)=s+a1x+a2x 2 +…+a k x k

[0077] Among them, f(x) represents the dependent variable, x represents the independent variable, k represents the order of the polynomial, a1, a2,…, a k They represent the 1st, 2nd, …, kth different fragmentation processing polynomial coefficients respectively, and s represents a decimal number.

[0078] In an embodiment of the present invention, obtaining a plurality of fragmented encryption keys according to the fragmentation processing polynomial includes:

[0079] The converted decimal number is divided into at least 2k key fragments, and each key fragment is brought into the fragmentation processing polynomial to obtain the target dependent variable corresponding to each key fragment;

[0080] For any key fragment, the key fragment and the target dependent variable are combined into a fragmented encryption key to obtain multiple fragmented encryption keys.

[0081] By obtaining multiple fragmented encryption keys, only k+1 fragmented encryption keys can be obtained during decryption to reconstruct the fragmented processing polynomial and realize decryption of the symmetric encryption key, thereby realizing decryption of the digital image data.

[0082] In an embodiment of the present invention, the encrypted digital image data is randomly transmitted to two first edge storage nodes through the main storage node, and each fragmented encryption key is transmitted to a different second edge storage node, including:

[0083] Randomly determine two first edge storage nodes through the main storage node, randomly transmit the encrypted digital image data to the two first edge storage nodes, and set the remaining edge storage nodes as second edge storage nodes; wherein the number of the second edge storage nodes is greater than 2k;

[0084] 2k second edge storage nodes are randomly determined, and each fragmented encryption key is transmitted to a different second edge storage node.

[0085] In an embodiment of the present invention, it also includes:

[0086] Obtaining a data query request input by a user; wherein the data query request includes at least one item of patient information;

[0087] According to the data query request, searching the corresponding target patient information, target data storage address and target key storage address through the main storage node;

[0088] According to the target data storage address and the target key storage address, the encrypted digital image data and the fragmented encryption key are obtained, and encrypted and transmitted through the public key of the user device.

[0089] The encrypted digital image data can be obtained according to the target data storage address, and at least k+1 fragmented encryption keys can be obtained according to the target key storage address. After the encrypted digital image data and at least k+1 fragmented encryption keys are transmitted to the user device, the data can be decrypted and viewed, thereby achieving comprehensive data security protection.

[0090] In the embodiment of the present invention, while obtaining the data query request input by the user, the following steps are also included:

[0091] Collect network traffic features, and use a pre-deployed deep learning model to detect the network traffic features to obtain network security detection results; wherein the network security detection results include network security or network intrusion;

[0092] When the network security detection result shows that the network has been invaded, a network security protection operation is performed to ensure the storage security of the digital image data.

[0093] Network security protection operations can be to block the user's IP for a certain period of time and prohibit the user from accessing data. If the number of attacks is too high, the port can be blocked for a certain period of time to avoid data leakage.

[0094] like Figure 2 As shown in the figure, the pre-deployment method of the deep learning model includes:

[0095] S201, initializing a population of the algorithm using a uniform initialization strategy; wherein each individual in the population is composed of all or part of the hyperparameters of the deep learning model;

[0096] For the d-th dimension parameter, randomly generate a parameter x between the upper and lower limits 1,d ; where x 1,d represents the d-th dimension parameter of the initial individual;

[0097] Based on the initial individual, the subsequent individuals are generated as follows:

[0098]

[0099] Among them, x n,d represents the d-th dimension parameter of the n-th individual, n=1,2,…,M, M represents the total number of individuals to be generated, rand1 represents a random number between (0,1), x n+1,d represents the d-th dimension parameter of the n+1th individual, γ d,min represents the lower limit of the d-th dimension parameter, γ d,max represents the upper limit of the d-th dimension parameter, α represents the segmentation parameter, and generally α is set to (γ d,min +γ d,max ) / 2.

[0100] By improving the initialization method, all individuals can be distributed more evenly in the solution space during initial training, which can effectively improve the convergence ability and speed of the algorithm. Compared with the chaotic mapping method, it can be more evenly distributed in the entire data, avoiding the technical problem of poor distribution of chaotic mapping in local parts.

[0101] It is worth noting that a random initialization strategy can also be used to initialize the population.

[0102] S202, sequentially performing exploration phase spatial search, development phase spatial search, encirclement phase spatial search, and escape phase spatial search on the population to obtain a population after multi-stage spatial search;

[0103] S202.1. Exploration phase space search, including:

[0104]

[0105] in, represents the mth individual in the tth training process, m = 1, 2, ..., M, M represents the total number of individuals in the population, rand2 represents a random number between (0, 1), rand3 represents a random number between (0, 1), Indicates except Random individuals other than Represents the individual after the spatial search in the exploration phase

[0106] Through spatial search in the exploration phase, the information of the two individuals can be mixed, improving the algorithm's local exploration capabilities and finding a better position within its search range.

[0107] S202.2. Development phase spatial search, including:

[0108] Based on the population after the spatial search in the exploration phase, the loss function value corresponding to each individual is obtained, and all individuals are arranged in order from small to large according to the loss function value. The weight coefficient corresponding to the individual for information interaction is obtained by sorting:

[0109]

[0110] in, Represents the weight coefficient corresponding to the i-th individual after sorting in the t-th training process.

[0111] Based on the weight coefficients corresponding to the individuals used for information interaction, the useful information of the population is obtained as follows:

[0112]

[0113] in, Indicates useful information of the population, The dimensions are the same as those of other individuals. Represents the i-th individual after sorting.

[0114] Useful information about the population Based on, the first update amount and the second update amount are determined as:

[0115]

[0116] in, Indicates the first update amount The d-th dimension parameter obeys (0, Cov i) is a normal distribution with a total parameter dimension of D, Cov i represents the intermediate parameter, T represents the transpose, express and The number product of represents the second update amount, represents the optimal individual in the t-th training process, represents the individual after the k-th exploration phase spatial search, k = 1, 2, …, M.

[0117] The individual update after the k-th exploration phase space search is:

[0118]

[0119] in, represents the individual after the spatial search in the development phase

[0120] By establishing a probability model to represent the relationship between individuals, the algorithm's search ability and local convergence are effectively improved. Then, by introducing the normal distribution, useful information can be learned more effectively, and with the help of learning the information of the best individual, it can not only ensure the algorithm's local search ability, but also effectively prevent the algorithm from falling into the local optimum too early.

[0121] S202.3. Encirclement phase spatial search, including:

[0122]

[0123] β=(ba)*rand4*e (tmax-t) / tmax

[0124] in, represents the individual after the h-th development stage space search, represents the individual after the spatial search in the encirclement phase β represents an adaptive adjustment parameter, a represents a first constant (set to 0.5 in this embodiment), b represents a second constant (set to 1 in this embodiment), rand4 represents a random number between (0, 1), and tmax represents a preset maximum number of training times.

[0125] In the early stages of algorithm iteration, in order to increase the diversity of the population, a larger adaptive adjustment parameter is used to generate a new solution. As the number of iterations increases, the algorithm will gradually tend toward the optimal solution, so it is necessary to gradually reduce the impact on the new solution, so as to achieve the optimal position of the surround search and improve the convergence accuracy of the algorithm.

[0126] S202.4. Escape phase space search, including:

[0127]

[0128] Among them, N(0,1) represents the standard Gaussian distribution function, represents the d-th dimension parameter of the individual after the q-th encirclement phase space search, Indicates the escape phase after the space search

[0129] By introducing Gaussian distribution, the individual's ability to mutate can be improved, so that the individual has a certain probability of escaping from its original position, which can effectively prevent the algorithm from falling into the local optimal solution and improve the algorithm's search ability.

[0130] Optionally, after each search, each individual may be subjected to out-of-bounds processing to ensure the effectiveness of the algorithm.

[0131] S203, determine whether the population after the multi-stage spatial search meets the iteration end condition. If so, select the best individual from the population after the multi-stage spatial search, and preset the deep learning model with the best individual, otherwise return to the multi-stage search step.

[0132] The pre-deployment method of the deep learning model provided by the embodiment of the present invention can effectively improve the hyperparameter optimization speed, avoid the algorithm from falling into the local optimum, and ultimately improve the ability of network intrusion detection and improve the data security protection effect.

[0133] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0134] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0135] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0136] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0137] A person of ordinary skill in the art can understand that all or part of the steps in realizing the above-mentioned facts and methods can be completed by instructing the relevant hardware through a program, and the program involved or the program described can be stored in a computer-readable storage medium. When the program is executed, it includes the following steps: At this time, the corresponding method steps are derived, and the storage medium can be ROM / RAM, a disk, an optical disk, etc.

[0138] The specific implementation methods described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation method of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A distributed storage method for digital image data, characterized in that: include: A distributed storage system is constructed using a cluster server; wherein the distributed storage system includes a main storage node and an edge storage node; Obtaining a data image data storage request through a primary storage node, and verifying the data image data storage request to obtain a request verification result; wherein the request verification result includes verification passed or verification failed; the data image data storage request includes patient information and digital image data; The digital image data is encrypted by using a fragmented encryption method through a main storage node to obtain the encrypted digital image data and a plurality of fragmented encryption keys; The encrypted digital image data is randomly transmitted to two first edge storage nodes through the main storage node, and each fragmented encryption key is transmitted to a different second edge storage node; The data storage address is returned through the first edge storage node and the key storage address is returned through the second edge storage node, and the patient information, the data storage address and the key storage address are associated and stored through the main storage node.

2. The distributed storage method for digital image data according to claim 1, characterized in that: Cluster servers are used to build a distributed storage system, including: Based on the cluster servers, determine the sum of the communication distances between each server and all other servers; According to the sum of the communication distances between each server and all other servers, the server with the smallest sum of communication distances is determined as the main storage node, and the other servers are determined as edge storage nodes.

3. The distributed storage method for digital image data according to claim 2, characterized in that: Obtaining a data image data storage request through a primary storage node, and verifying the data image data storage request to obtain a request verification result, including: Obtaining a data image data storage request through a main storage node, and obtaining patient information to be stored and digital image data to be stored; Performing online verification on the name, gender and ID card number in the stored patient information to obtain a basic verification result; wherein the basic verification result includes verification success or verification failure; When the basic verification result is verification failure, determining the request verification result is verification failure; When the basic verification result is a successful verification, a missing verification is performed on the inspection information and the inspection sequence information to obtain a missing verification result; wherein the missing verification result includes a successful verification or a failed verification; When the missing verification result is a successful verification, the request verification result is determined to be a passed verification, otherwise the request verification result is determined to be a failed verification.

4. The distributed storage method for digital image data according to claim 3, characterized in that: The digital image data is encrypted by the main storage node using a fragmented encryption method to obtain the encrypted digital image data and multiple fragmented encryption keys, including: The digital image data is encrypted using a symmetric encryption algorithm to obtain the encrypted digital image data and the corresponding symmetric encryption key; Based on the symmetric encryption key, a fragmentation processing polynomial is generated, and a plurality of fragmentation encryption keys are obtained according to the fragmentation processing polynomial.

5. The distributed storage method for digital image data according to claim 4, characterized in that: Based on the symmetric encryption key, the fragmentation processing polynomial is generated as follows: The symmetric encryption key is converted into a decimal number, and a fragmentation processing polynomial is constructed according to the converted decimal number: f(x)=s+a1x+a2x 2 +…+a k x k Among them, f(x) represents the dependent variable, x represents the independent variable, k represents the order of the polynomial, a1, a2,…, a k They represent the 1st, 2nd, …, kth different fragmentation processing polynomial coefficients respectively, and s represents a decimal number.

6. The distributed storage method for digital image data according to claim 5, characterized in that: Acquiring a plurality of fragmented encryption keys according to the fragmentation processing polynomial includes: The converted decimal number is divided into at least 2k key fragments, and each key fragment is brought into the fragmentation processing polynomial to obtain the target dependent variable corresponding to each key fragment; For any key fragment, the key fragment and the target dependent variable are combined into a fragmented encryption key to obtain multiple fragmented encryption keys.

7. The distributed storage method for digital image data according to claim 6, characterized in that: The encrypted digital image data is randomly transmitted to two first edge storage nodes through the main storage node, and each fragmented encryption key is transmitted to a different second edge storage node, including: Randomly determine two first edge storage nodes through the main storage node, randomly transmit the encrypted digital image data to the two first edge storage nodes, and set the remaining edge storage nodes as second edge storage nodes; wherein the number of the second edge storage nodes is greater than 2k; 2k second edge storage nodes are randomly determined, and each fragmented encryption key is transmitted to a different second edge storage node.

8. The distributed storage method for digital image data according to claim 7, characterized in that: Also includes: Obtaining a data query request input by a user; wherein the data query request includes at least one item of patient information; According to the data query request, searching the corresponding target patient information, target data storage address and target key storage address through the main storage node; According to the target data storage address and the target key storage address, the encrypted digital image data and the fragmented encryption key are obtained, and encrypted and transmitted through the public key of the user device.

9. The distributed storage method for digital image data according to claim 8, characterized in that: When obtaining the data query request entered by the user, it also includes: Collect network traffic features, and use a pre-deployed deep learning model to detect the network traffic features to obtain network security detection results; wherein the network security detection results include network security or network intrusion; When the network security detection result shows that the network has been invaded, a network security protection operation is performed to ensure the storage security of the digital image data.

10. The distributed storage method for digital image data according to claim 9, characterized in that: Pre-deployment methods for deep learning models, including: A population of algorithms is initialized using a uniform initialization strategy, where each individual in the population consists of all or part of the hyperparameters of the deep learning model; Perform exploration phase spatial search, development phase spatial search, encirclement phase spatial search and escape phase spatial search on the population in sequence to obtain the population after multi-stage spatial search; Determine whether the population after the multi-stage spatial search meets the iteration end condition. If so, select the best individual from the population after the multi-stage spatial search and preset the deep learning model with the best individual. Otherwise, return to the multi-stage search step.

Images