Image desensitization testing method, device and computer-readable storage medium

Abstract

The present application relates to an image desensitization test method, device and computer-readable storage medium, which belongs to the technical field of big data, and the method includes: obtaining multiple data sets to be tested; wherein the image types of the image data contained in the same data set to be tested are the same; according to the data size and image type of each data set to be tested, the final computing node of each data set to be tested is determined from multiple cloud computing nodes, so that the final computing node performs an image desensitization test on the assigned data set to be tested, and generates an image desensitization test result for each data set to be tested. The present application has the effect of improving the efficiency of large-scale desensitization testing.

Description

Technical Field

[0001] The present application relates to the technical field of big data, and in particular to an image desensitization testing method, device and computer-readable storage medium. Background Art

[0002] With the rapid development of big data and the Internet, network information has increased, especially image information involving user privacy information. Therefore, network information security faces huge challenges.

[0003] At present, image desensitization is usually used to protect user privacy. Image desensitization refers to erasing or covering images based on desensitization rules for certain sensitive information, thereby achieving reliable protection of sensitive privacy data.

[0004] Usually, after desensitization of images, it is necessary to check and compare the desensitized images to verify the desensitization results. Different types of images have different applicable desensitization test algorithms, which are generally related to the desensitized content, location, scene, etc.

[0005] The current desensitization test system generally tests a specific type of desensitized image. If multiple different types of desensitized images are tested, multiple image desensitization tests need to be performed according to the image type to complete the test. In this way, not only multiple desensitization test algorithms need to be integrated into the computer, but also higher requirements are placed on the computer's computing power.

[0006] Therefore, there is an urgent need for an image desensitization testing technology that can simultaneously perform desensitization testing on a dataset containing images of different image types. Summary of the invention

[0007] In order to improve the efficiency of large-scale desensitization testing, the present application provides an image desensitization testing method, device and computer-readable storage medium.

[0008] In the first aspect, the present application provides an image desensitization testing method, which adopts the following technical solution:

[0009] An image desensitization testing method, comprising:

[0010] Acquire multiple data sets to be tested; wherein the image data contained in the same data set to be tested have the same image type;

[0011] According to the data size and image type of each data set to be tested, the final computing node of each data set to be tested is determined from multiple cloud computing nodes, so that the final computing node performs an image desensitization test on the assigned data set to be tested, and generates an image desensitization test result for each data set to be tested.

[0012] By adopting the above technical solution, cloud computing resources can be flexibly allocated according to the data size and image type of the test data set, so as to support desensitization testing of various image types. It has strong scalability and enables each cloud computing node to perform desensitization testing on the test data set that matches its processing capacity, thereby ensuring the test accuracy and efficiency of the test data set to a certain extent.

[0013] Optionally, the number of the cloud computing nodes is not less than the number of the data sets to be tested; and determining the final computing node of each data set to be tested from the plurality of cloud computing nodes according to the data size and image type of each data set to be tested comprises:

[0014] Obtaining desensitization type information of each of the cloud computing nodes; wherein the desensitization type information includes at least one desensitization type; and the desensitization type corresponds to the image type one by one;

[0015] Matching at least one to-be-allocated computing node for each of the data sets to be tested based on the image type of the data sets to be tested and the desensitization type information;

[0016] Determine whether there is a test data set that matches only one computing node to be assigned;

[0017] If there is a data set to be tested that matches only one computing node to be allocated, the computing node to be allocated that matches the data set to be tested is determined as the final computing node of the data set to be tested, and the final computing node is recorded as the allocated computing node, and the step of determining whether there is a data set to be tested that matches only one computing node to be allocated is performed until there is no data set to be tested that matches only one computing node to be allocated;

[0018] If there is no data set to be tested that matches only one computing node to be allocated, the data set to be tested that matches multiple computing nodes to be allocated is determined as the target data set to be tested, and the final computing node of the target data set to be tested is determined according to the data size of the target data set to be tested and the computing resource information of the computing node to be allocated.

[0019] By adopting the above technical solutions, reasonable matching and optimized allocation can be achieved under limited cloud computing resources, the performance of each cloud computing node can be fully utilized, and the overall performance and efficiency can be improved.

[0020] Optionally, determining the final computing node of the target data set to be tested according to the data size of the target data set to be tested and the computing resource information of the computing node to be allocated includes:

[0021] Sorting all target test data sets according to the data size of the test data sets to obtain a first sorting result;

[0022] Obtain computing resource information of each computing node to be allocated; wherein the computing resource information includes memory, processing speed and network bandwidth;

[0023] Calculate a score value of each to-be-allocated computing node according to the computing resource information;

[0024] According to the score value, sorting the multiple to-be-allocated computing nodes corresponding to each of the target data sets to be tested, to obtain a second sorting result corresponding to the target data sets to be tested;

[0025] According to the first sorting result, the target data sets to be tested are selected in sequence; for the selected target data sets to be tested, according to the second sorting result corresponding to the selected target data sets to be tested, the final computing node of the selected target data sets to be tested is determined, and the final computing node is recorded as the allocated computing node; until the final computing node of each target data set to be tested is determined.

[0026] By adopting the above technical solution, the data size of the data set to be tested and the computing resource information of the computing node to be allocated are comprehensively considered, and a cloud computing node is allocated to each data set to be tested according to the first sorting result of each data set to be tested and the second sorting result of each cloud computing node. Among them, for the data set to be tested with a large amount of data, a computing node with better comprehensive performance is selected for processing to improve the overall processing capability and efficiency.

[0027] Optionally, after respectively determining the final computing node of each of the data sets to be tested from a plurality of cloud computing nodes according to the data size and image type of each of the data sets to be tested, the method further includes:

[0028] Obtaining a preset time threshold for each data set to be tested based on the data size of the data set to be tested;

[0029] Based on the processing speed, the network bandwidth and the data size of the data set to be tested, predict the test time of each of the cloud computing nodes processing the assigned data set to be tested;

[0030] Determine whether there is a cloud computing node whose test time is greater than the preset time threshold;

[0031] If there is a computing node whose test time is greater than the preset time threshold, the to-be-tested data set corresponding to the cloud computing node whose test time is greater than the preset time threshold is used as the to-be-divided data set, and the cloud computing node whose test time is greater than the preset time threshold is used as the first node;

[0032] Determine whether there is a cloud computing node that processes the image type of the data set to be divided and the test time is less than the preset time threshold;

[0033] If there is a cloud computing node that tests the image type of the data set to be divided and the test time is less than the preset time threshold, the cloud computing node is used as the second node; the data set to be divided is divided based on the test time of the first node and the preset time threshold to obtain a first divided data set and a second divided data set;

[0034] Determining whether the second node can process the second divided data set within the preset time threshold;

[0035] If the second node can process the second partitioned data set within the preset time threshold, the second partitioned data set is allocated to the second node.

[0036] By adopting the above technical solution, the current processing performance of each cloud computing node can be further fully utilized to ensure overall processing efficiency and stability.

[0037] Optionally, matching at least one to-be-allocated computing node for each of the data sets to be tested based on the image type of the data sets to be tested and the desensitization type information includes:

[0038] For any of the data sets to be tested, a cloud computing node corresponding to a desensitization type that corresponds one-to-one to the image type of the data set to be tested is searched from the multiple cloud computing nodes, and the found cloud computing node is used as a computing node to be allocated for the data set to be tested.

[0039] Optionally, also include:

[0040] Obtaining the image desensitization test result returned by each of the final computing nodes;

[0041] The desensitization test results of each image type are analyzed based on the image desensitization test results.

[0042] By adopting the above technical solution, the test effects of various image types can be evaluated according to the desensitization test results of the data set to be tested, so as to optimize and adjust the corresponding desensitization algorithm.

[0043] In a second aspect, the present application provides an image desensitization testing device, which adopts the following technical solution:

[0044] An image desensitization testing device, comprising:

[0045] A first acquisition module is used to acquire multiple data sets to be tested; wherein the image data contained in the same data set to be tested have the same image type;

[0046] The first determination module is used to determine the final computing node of each data set to be tested from multiple cloud computing nodes according to the data size and image type of each data set to be tested, so that the final computing node performs an image desensitization test on the assigned data set to be tested, and generates an image desensitization test result for each data set to be tested.

[0047] In a third aspect, the present application provides a control node device, which adopts the following technical solution:

[0048] A control node device comprises a processor, wherein the processor is coupled to a memory;

[0049] The processor is used to execute the computer program stored in the memory, so that the electronic device executes the method described in any one of the first aspects.

[0050] In a fourth aspect, the present application provides a cloud service platform, which adopts the following technical solutions:

[0051] A cloud service platform comprises a control node device, a cloud storage node, a cloud network node and a plurality of cloud computing nodes as described in the third aspect.

[0052] In a fifth aspect, the present application provides a computer-readable storage medium, which adopts the following technical solution:

[0053] A computer-readable storage medium storing a computer program that can be loaded by a processor and execute the image desensitization testing method as described in any one of the first aspects.

[0054] By adopting the above technical solution, the processor loads and executes the computer program stored in the computer-readable storage medium, and flexibly allocates cloud computing resources according to the data size and image type of the test data set, so as to support desensitization testing of various image types, and has strong scalability. In addition, each cloud computing node performs desensitization testing on the test data set that matches its processing capability, thereby ensuring the test accuracy and efficiency of the test data set to a certain extent. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 It is a structural block diagram of a cloud service platform in an embodiment of the present application.

[0056] Figure 2 It is a flow chart of an image desensitization testing method applied to a control node device according to an embodiment of the present application.

[0057] Figure 3 It is a flowchart of steps S1021 to S1025 of an embodiment of the present application.

[0058] Figure 4 It is a flow chart of steps Sa to Se of an embodiment of the present application.

[0059] Figure 5 It is a structural block diagram of an image desensitization testing device according to an embodiment of the present application.

[0060] Figure 6 It is a flow chart of an image desensitization testing method applied to a cloud computing node in an embodiment of the present application.

[0061] Figure 7 It is a structural block diagram of a control node device in an embodiment of the present application. DETAILED DESCRIPTION

[0062] The present application is further described in detail below in conjunction with the accompanying drawings.

[0063] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0064] In addition, the term "and / or" in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article, unless otherwise specified, generally means that the associated objects before and after are in an "or" relationship.

[0065] Figure 1 A cloud service platform for image desensitization testing is provided in the embodiment of the present application. Figure 1 As shown, the platform includes a control node device, a cloud storage node, a cloud network node and multiple cloud computing nodes.

[0066] The above nodes realize resource pooling through virtualization technology to form a cloud platform. Cloud computing nodes can be configured and managed according to actual needs to achieve efficient and flexible cloud computing services. The control node is the core component of the cloud service platform, responsible for managing and scheduling other nodes. The cloud network node provides network connection and management services to ensure the stability and security of internal communications of the cloud service platform. The cloud storage node is responsible for data storage and management, providing high-performance and scalable storage services.

[0067] In this embodiment, the number of cloud computing nodes can be configured according to the image type, the data of the cloud computing nodes is not less than the number of image types, and at least one desensitization test rule set according to the image type is pre-configured in the cloud computing node. The desensitization test rule corresponds one-to-one to the desensitization type, and the desensitization type corresponds one-to-one to the image type.

[0068] certainly, Figure 1 Only the relationship between the cloud storage nodes, cloud network nodes and control nodes is shown, and the number of cloud storage nodes and cloud network nodes can be set according to specific circumstances.

[0069] Figure 2 An image desensitization testing method provided in an embodiment of the present application is applied to a control node device, which can be a server or a loosely coupled server cluster composed of two or more servers, but is not limited to this.

[0070] like Figure 2 As shown, the main process of the method includes the following steps S101-S102:

[0071] Step S101, obtaining multiple data sets to be tested;

[0072] In this embodiment, each data set to be tested includes at least one desensitized image and a non-desensitized image corresponding to the desensitized image. The image types of the image data included in these data sets to be tested may be the same or different, but the image types of the image data included in each data set to be tested need to be the same.

[0073] Since the content to be desensitized is different for different images, for example, the desensitized content may be face information, license plate information, official seal information, ID card information, face information, address information, password information, etc., the image type can be determined according to the desensitized content. For example, the image type can be a courier image with the desensitized content of sender / recipient information, an ID card image with the desensitized content of ID card information, a car image with the desensitized content of car license plate, and a contract image with the desensitized content of official seal / signature / enterprise or personal information.

[0074] It should be noted that the image type may be directly input by the user. For example, after determining the data set to be tested, the image type option corresponding to the data set to be tested is selected through the image type selection interface.

[0075] Step S102, according to the data size and image type of each data set to be tested, determine the final computing node of each data set to be tested from multiple cloud computing nodes, so that the final computing node performs an image desensitization test on the assigned data set to be tested, and generates an image desensitization test result for each data set to be tested.

[0076] Specifically, Figure 3 As shown, step S102 includes the following sub-steps:

[0077] Step S1021, obtaining desensitization type information of each cloud computing node; wherein the desensitization type information includes at least one desensitization type; and the desensitization type corresponds to the image type one by one;

[0078] In step S1021, each cloud computing node sets at least one desensitization test rule in advance according to the image type, and the desensitization test rule is associated with the desensitization type, that is, each desensitization type corresponds to a desensitization test rule. For example, some cloud computing nodes are provided with a desensitization test rule of one desensitization type, and some cloud computing nodes are provided with desensitization test rules of multiple desensitization types. The configuration personnel can set it according to the performance of the computer, so that the desensitization test rules set in all cloud computing nodes can cover all image types to be tested.

[0079] Step S1022, matching at least one to-be-allocated computing node for each data set to be tested based on the image type and desensitization type information of the data set to be tested;

[0080] Specifically, for any data set to be tested, a cloud computing node corresponding to a desensitization type that corresponds one-to-one to the image type of the data set to be tested is searched from multiple cloud computing nodes, and the found cloud computing node is used as the computing node to be allocated for the data set to be tested.

[0081] For example, there are existing test data sets 1, 2 and 3, the image type of test data set 1 is image type 1, the image type of test data set 2 is image type 2, and the image type of test data set 3 is image type 3. The cloud service platform is configured with three cloud computing nodes, namely cloud computing node 1, cloud computing node 2 and cloud computing node 3. Among them, the desensitization type information configured for cloud computing node 1 is desensitization type 2 and desensitization type 3, the desensitization types configured for cloud computing node 2 are desensitization type 1, desensitization type 2 and desensitization type 3, and the desensitization type configured for cloud computing node 3 is desensitization type 3. Then, the computing node to be allocated that matches the test data set 1 is cloud computing node 2, the computing nodes to be allocated that match the test data set 2 are cloud computing node 1 and cloud computing node 2, and the computing nodes to be allocated that match the test data set 3 are cloud computing node 2 and cloud computing node 3.

[0082] Step S1023, determine whether there is a test data set that matches only one computing node to be assigned; if so, proceed to step S1024; if not, proceed to step S1025.

[0083] Continuing with the example in step S1022, the only computing node to be allocated that matches the test data set 1 is the cloud computing node 2, so the test data set 1 is a test data set that only matches one computing node to be allocated, while the test data sets 2 and 3 both match multiple computing nodes to be allocated.

[0084] Step S1024, determining the to-be-allocated computing node that matches the to-be-tested data set as the final computing node of the to-be-tested data set, and recording the final computing node as the allocated computing node, and executing the step of determining whether there is a to-be-tested data set that matches only one to-be-allocated computing node, until there is no to-be-tested data set that matches only one to-be-allocated computing node;

[0085] Step S1025 , determining the test data set matching multiple computing nodes to be allocated as the target test data set, and determining the final computing node of the target test data set according to the data size of the target test data set and the computing resource information of the computing nodes to be allocated.

[0086] Continuing with the example in step S1022, cloud computing node 2 is assigned to the test data set 1, determined as the final computing node, and the final computing node is recorded as the assigned computing node, and step S1033 is repeated. The test data set 2 and the test data set 3 are both test data sets that only match one to-be-assigned computing node. In this case, the to-be-assigned computing node is directly assigned and determined as the final computing node, that is, cloud computing node 1 is assigned to the test data set 2, and cloud computing node 3 is matched to the test data set 3.

[0087] In this embodiment, if there is no test data set that matches only one computing node to be assigned, it means that the test data set for which the final computing node has not been determined can be desensitized by multiple cloud computing nodes. Then the test data set that matches multiple computing nodes to be assigned is determined as the target test data set, and the final computing node of the target sub-data is determined based on the data size of the target test data set and the computing resource information of the computing node to be assigned.

[0088] This ensures that with limited computing resources, the performance of each cloud computing node is fully utilized, the overall performance and efficiency are improved, and reasonable matching and optimized allocation are achieved to improve the overall processing capacity and efficiency.

[0089] Further, such as Figure 4 As shown, as an optional implementation method of determining the final computing node of the target data set to be tested according to the data size of the target data set to be tested and the computing resource information of the computing node to be allocated, step S1025 includes:

[0090] Step Sa, sorting all target test data sets according to the data size of the test data sets to obtain a first sorting result;

[0091] Specifically, the first sorting result may be sorted from large to small according to the data size of the target data set to be tested.

[0092] Step Sb, obtaining computing resource information of each computing node to be allocated; wherein the computing resource information includes memory, processing speed and network bandwidth;

[0093] Step Sc, calculating the score value of each computing node to be allocated according to the computing resource information;

[0094] Specifically, you can set respective weights for memory, processing speed, and network bandwidth, calculate the memory usage (percentage), processing speed ratio, and network bandwidth ratio, and then perform the weighted sum to obtain the score value of the computing node to be allocated.

[0095] Step Sd, sorting the multiple computing nodes to be allocated corresponding to each target data set to be tested according to the score value, to obtain a second sorting result corresponding to the target data set to be tested;

[0096] Step Se, select the target data set to be tested in sequence according to the first sorting result; for the selected target data set to be tested, determine the final computing node of the selected target data set to be tested according to the second sorting result corresponding to the selected target data set to be tested, and record the final computing node as the allocated computing node; until the final computing node of each target data set to be tested is determined.

[0097] In this optional implementation, if the target data set to be tested 1 matches the to-be-allocated computing nodes 1 and 3, and the target data set to be tested 1 matches the to-be-allocated computing nodes 1, 2, and 3, and the scores of the to-be-allocated computing nodes are Node 2 to be allocated> Node 3 to be allocated> Node 1 to be allocated, then the multiple to-be-allocated computing nodes are sorted according to the score values ​​of the nodes to be allocated. The second sorting result corresponding to the target data set to be tested 1 is Node 3 to be allocated and Node 1 to be allocated; the second sorting result of the target data set to be tested 2 is Node 2 to be allocated, Node 3 to be allocated, and Node 1 to be allocated.

[0098] First, a target data set to be tested is selected according to the first sorting result, and a to-be-allocated computing node with the highest score in the second sorting result corresponding to the target data set to be tested is selected as the final computing node, and the final computing node is recorded as an allocated node.

[0099] Afterwards, determine whether there is a test data set in the remaining target test data sets that only contains one to-be-allocated computing node; if so, determine the to-be-allocated computing node contained in the target test data set as the final computing node of the target test data set, and record the final computing node as the allocated computing node.

[0100] Finally, the step of selecting target data sets to be tested in sequence according to the first sorting result is performed until the final computing node of each target data set to be tested is determined.

[0101] By comprehensively considering the data size of the data set to be tested and the computing resource information of the computing node to be allocated, cloud computing nodes are allocated to the data set to be tested according to the first sorting result of the data set to be tested and the second sorting result of the cloud computing node. For the data set to be tested with a larger data size, cloud computing nodes with better comprehensive performance are used for processing, thereby improving the overall processing capability and efficiency.

[0102] In some embodiments, after respectively determining the final computing node of each of the data sets to be tested from a plurality of cloud computing nodes according to the data size and image type of each of the data sets to be tested, the method further comprises:

[0103] First, a preset time threshold of each data set to be tested is obtained based on the data size of the data set to be tested; wherein the preset time threshold is pre-set according to the historical execution status and the data size of the data set to be tested, and can be directly obtained and called.

[0104] Then, based on the processing speed, network bandwidth and the data size of the data set to be tested, the test time of each cloud computing node to process the assigned data set to be tested is predicted. The time required for the current test can be evaluated based on the historical test results of the cloud computing node.

[0105] When the data size of the data set to be tested, that is, the data volume, is large, the cloud computing node performs desensitization testing on it slowly. Therefore, it is necessary to further divide the data set to be tested corresponding to the cloud computing node whose test time is greater than the preset time threshold to check whether there are other allocated or unallocated computing nodes that can process it.

[0106] If there is a computing node whose test time is greater than the preset time threshold, the to-be-tested data set corresponding to the cloud computing node whose test time is greater than the preset time threshold is used as the to-be-divided data set, and the cloud computing node whose test time is greater than the preset time threshold is used as the first node.

[0107] Afterwards, it is necessary to determine whether there is a cloud computing node that processes the image type of the data set to be divided and whose test time is less than a preset time threshold; it should be noted that the cloud computing node can be an allocated computing node or an unallocated computing node.

[0108] If there is a cloud computing node that tests the image type of the data set to be divided and the test time is less than a preset time threshold, the cloud computing node is used as the second node; the data set to be divided is divided based on the test time of the first node and the preset time threshold to obtain a first divided data set and a second divided data set.

[0109] If there is no cloud computing node that tests the image type of the data set to be divided, and the test time is less than the preset time threshold, the final computing node previously assigned will first perform a desensitization test on the data set to be divided. If a cloud computing node that can process the image type of the data set to be divided appears subsequently, some image data in the data set to be divided that has not been desensitized will be sent to the cloud computing node for desensitization testing.

[0110] Finally, determine whether the second node can process the second partitioned data set within a preset time threshold; if the second node can process the second partitioned data set within the preset time threshold, assign the second partitioned data set to the second node and assign the first partitioned data set to the first node.

[0111] It should be noted that the second node can be one or more, as long as it meets the two conditions of being able to process the image type of the data set to be divided and the test time is less than the preset time threshold. Therefore, there may be multiple second nodes that meet the above conditions. If there are unallocated computing nodes among these second nodes, the unallocated computing nodes are preferentially used as the second nodes. If there are no unallocated computing nodes among these second nodes, the allocated computing nodes that meet the above conditions are used as the second nodes.

[0112] In this embodiment, the unallocated computing nodes are cloud computing nodes that are not allocated any data set to be tested; and the allocated computing nodes are cloud computing nodes that are allocated the data set to be tested.

[0113] In addition, when dividing the data set to be divided into the first divided data set and the second divided data set, the division may be performed according to the test time, or according to the processing capability of the cloud computing node or in other ways, and is not limited thereto.

[0114] After the above node allocation, one data set to be tested may be tested by multiple cloud computing nodes at the same time, and one cloud computing node may also test all or part of the image data of multiple data sets to be tested in sequence.

[0115] In some embodiments, after obtaining the image desensitization test results returned by the final computing node, these image desensitization test results are stored, and the desensitization effects of each image type can be regularly summarized and evaluated based on the stored image desensitization test results, so as to optimize and adjust the corresponding desensitization test rules.

[0116] In this embodiment, after the cloud computing node completes the desensitization test on the assigned data set to be tested, the desensitization test result of the image of the data set to be tested is sent to the control node device, and the desensitization test result of the image of the data set to be tested includes image type, image desensitization pass rate, image desensitization abnormality, etc. The control node device analyzes the desensitization test result of the data set to be tested according to the image type, image desensitization pass rate, image desensitization abnormality, number of images, etc. of each data set to be tested, and comprehensively analyzes the desensitization test result to improve the desensitization test rules for different image types.

[0117] This method allocates a specific cloud computing node to each test data set for image desensitization testing, and processes different test data sets in parallel, thereby improving the overall test efficiency. The scalability of the method is enhanced by flexibly allocating computing resources according to the data size and image type of the test data set, and each cloud computing node performs a desensitization test on the test data set that matches its processing capacity, thereby ensuring the test accuracy and efficiency of each test data set. At the same time, the method supports desensitization testing of multiple image types, and can obtain more comprehensive desensitization test results.

[0118] Figure 5 This is a structural block diagram of an image desensitization testing device 500 provided in an embodiment of the present application. Figure 5 As shown, the device 500 includes:

[0119] The first acquisition module 501 is used to acquire multiple data sets to be tested; wherein the image data contained in the same data set to be tested have the same image type;

[0120] The first determination module 502 is used to determine the final computing node of each data set to be tested from multiple cloud computing nodes according to the data size and image type of each data set to be tested, so that the final computing node performs an image desensitization test on the assigned data set to be tested and generates an image desensitization test result for each data set to be tested.

[0121] In this embodiment, the number of cloud computing nodes is not less than the number of data sets to be tested.

[0122] In some embodiments, the first determining module 502 includes the following submodules:

[0123] An acquisition submodule is used to acquire desensitization type information of each cloud computing node; wherein the desensitization type information includes at least one desensitization type; and the desensitization type corresponds to the image type one by one;

[0124] A matching submodule, used for matching at least one to-be-allocated computing node for each data set to be tested based on the image type and desensitization type information of the data set to be tested;

[0125] A judgment submodule is used to judge whether there is a test data set that matches only one computing node to be assigned;

[0126] A first determination submodule is used for, if there is a data set to be tested that matches only one computing node to be allocated, determining the computing node to be allocated that matches the data set to be tested as the final computing node of the data set to be tested, and recording the final computing node as the allocated computing node, and executing the step of determining whether there is a data set to be tested that matches only one computing node to be allocated, until there is no data set to be tested that matches only one computing node to be allocated;

[0127] The second determination submodule is used to determine the test data set that matches multiple computing nodes to be assigned as the target test data set if there is no test data set that matches only one computing node to be assigned, and determine the final computing node of the target test data set according to the data size of the target test data set and the computing resource information of the computing node to be assigned.

[0128] Furthermore, the second determination submodule is specifically used to sort all target data sets to be tested according to the data size of the data sets to be tested to obtain a first sorting result; obtain computing resource information of each computing node to be allocated; wherein the computing resource information includes memory, processing speed and network bandwidth; calculate the score value of each computing node to be allocated according to the computing resource information; sort the multiple computing nodes to be allocated corresponding to each target data set to be tested according to the score value, and obtain a second sorting result corresponding to the target data set to be tested; select the target data set to be tested in turn according to the first sorting result; for the selected target data set to be tested, determine the final computing node of the selected target data set to be tested according to the second sorting result corresponding to the selected target data set to be tested, and record the final computing node as the allocated computing node; until the final computing node of each target data set to be tested is determined.

[0129] Optionally, a matching submodule is specifically used to search, for any data set to be tested, from multiple cloud computing nodes for a cloud computing node corresponding to a desensitization type that corresponds one-to-one to the image type of the data set to be tested, and use the found cloud computing node as the computing node to be allocated for the data set to be tested.

[0130] In some embodiments, the device further comprises:

[0131] A second acquisition module is used to obtain a preset time threshold of each data set to be tested based on the data size of each data set to be tested, after respectively determining the final computing node of each data set to be tested from multiple cloud computing nodes according to the data size and image type of each data set to be tested;

[0132] A prediction module, used for predicting the test time of each cloud computing node processing the assigned data set to be tested based on the processing speed, network bandwidth and the data size of the data set to be tested;

[0133] A first judgment module is used to judge whether there is a cloud computing node whose test time is greater than a preset time threshold;

[0134] A second determination module is used to, if there is a computing node whose test time is greater than a preset time threshold, use the to-be-tested data set corresponding to the cloud computing node whose test time is greater than the preset time threshold as the to-be-divided data set, and use the cloud computing node whose test time is greater than the preset time threshold as the first node;

[0135] The second judgment module is used to judge whether there is a cloud computing node that processes the image type of the data set to be divided and the test time is less than a preset time threshold;

[0136] A partitioning module is determined, which is used to use the cloud computing node as the second node if there is a cloud computing node that tests the image type of the data set to be partitioned and the test time is less than a preset time threshold; the data set to be partitioned is partitioned based on the test time of the first node and the preset time threshold to obtain a first partitioned data set and a second partitioned data set;

[0137] A third judgment module, used to judge whether the second node can process the second divided data set within a preset time threshold;

[0138] The allocation module is configured to allocate the second divided data set to the second node if the second node can process the second divided data set within a preset time threshold.

[0139] The device also includes:

[0140] The third acquisition module is used to obtain the image desensitization test result returned by each final computing node;

[0141] The analysis module is used to analyze the desensitization test results of each image type based on the image desensitization test results.

[0142] Each functional module in the embodiment of the present application can be integrated together to form an independent unit, for example, integrated in a processing unit, or each module can exist physically separately, or two or more modules can be integrated to form an independent unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of a software functional unit. If the function is implemented in the form of a software functional module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or the part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including several instructions to enable an electronic device (which can be a personal computer, a server or a network device, etc.) or a processor (processor) to perform all or part of the steps of the methods of each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory, a random access memory, a magnetic disk or an optical disk.

[0143] The various variations and specific examples in the method provided in the embodiment of the present application are also applicable to an image desensitization testing device provided in the present embodiment. Through the above detailed description of the image desensitization testing method, those skilled in the art can clearly know the implementation method of the image desensitization testing device in the present embodiment. For the sake of brevity of the specification, it will not be described in detail here.

[0144] Correspondingly, an embodiment of the present application also provides an image desensitization testing method, which is applied to a cloud computing node. Figure 6 The figure shows a flow chart of an image desensitization test method applied to a cloud computing node. Figure 6 As shown, the main process of the method includes steps S201 to S204:

[0145] Step S201, obtaining a data set to be tested sent by a control node device;

[0146] In the implementation manner of the present application, the communication between the control node device and the cloud computing node can adopt symmetric encryption, asymmetric encryption and the like to ensure the security of the transmission of the data set to be tested. The cloud computing node decrypts the data set to be tested sent by the control node device according to a preset decryption algorithm. The data set to be tested contains at least one desensitized image and a non-desensitized image corresponding to the desensitized image.

[0147] Step S202, obtaining a desensitization test rule based on the image type of the data set to be tested;

[0148] In an embodiment of the present application, the desensitization test rules are set according to the image type. For example, one desensitization test rule is used for desensitization test of a fixed-format form containing identity information, one desensitization test rule is used for desensitization test of image data containing face images, and one desensitization test rule is used for desensitization test of image data divided into image data containing car images.

[0149] Step S203, performing an image desensitization test on the test data set based on the desensitization test rules to obtain an image desensitization test result;

[0150] In the implementation manner of the present application, specifically, step S203 includes:

[0151] Step S2031, for any desensitized image, according to the desensitized image, the non-desensitized image corresponding to the desensitized image and the desensitization test rule, determining at least one detection area of ​​the desensitized image and the non-desensitized image corresponding thereto;

[0152] In the implementation mode of the present application, according to the desensitization test rules, the desensitized image and the non-desensitized image are compared to determine at least one detection area of ​​the desensitized image and the corresponding non-desensitized image. For example, for an image of the type of express form, the detection area includes an area corresponding to the address content, an area corresponding to the mobile phone number information, and the like.

[0153] Step S2032, determining whether each detection area of ​​the desensitized image includes a desensitized erasure area;

[0154] In the implementation manner of the present application, taking the express delivery form in step S2031 as an example, the private information that needs to be desensitized is the address content, mobile phone number information, etc., then the area covered by the desensitization is the desensitized erase area.

[0155] Step S2033, if each detection area includes a desensitization erased area, then based on the content in the detection area and the content in the desensitization erased area corresponding to the detection area, determine the desensitization test result of the desensitized image;

[0156] In the implementation manner of the present application, the desensitization test result of the desensitized image includes that the desensitized image is a desensitization qualified image and the desensitized image is a desensitization unqualified image. For the desensitization unqualified image, it is necessary to continue to record the abnormal situation of the unqualified image.

[0157] Specifically, first, the cloud computing node detects the intersection-and-union ratio of each of the detection areas and the desensitized erased areas; the intersection-and-union ratio is the ratio of the intersection of the detection area and the desensitized erased area to the union of the detection area and the desensitized erased area; if the intersection-and-union ratio of each detection area in the desensitized image is not less than a preset intersection-and-union ratio, the desensitized image is determined to be a qualified desensitized image; if the intersection-and-union ratio of any detection area in the desensitized image is less than a preset intersection-and-union ratio, the content in the detection area in the non-desensitized image whose intersection-and-union ratio is less than the preset intersection-and-union ratio is identified.

[0158] In the implementation mode of the present application, the preset intersection-and-union ratio is determined based on the range of the detection area and the desensitized erased area in the historical desensitized qualified image. When the intersection-and-union ratio is not less than the preset intersection-and-union ratio, it means that the desensitized erased area covers a large area of ​​the detection area, so that the detection area can be directly judged as qualified for desensitization without identifying the content in the detection area, thereby reducing the test time and the use of computing resources.

[0159] When the calculated IoU is less than the preset IoU, it means that the coverage of the desensitized erasure area over the detection area does not meet the standard, and sensitive words may not be desensitized. In this case, it is necessary to identify the content in the detection area where the IoU is less than the preset IoU, and determine whether the desensitization of the detection area is qualified based on the content.

[0160] Afterwards, the detection area is divided into a plurality of target areas according to the content in the detection area; and weight values ​​are assigned to the target areas according to the positions and contents of the target areas.

[0161] In the implementation mode of the present application, the content of the detection area in the non-desensitized image can be feature extracted to obtain content information. The detection area is divided into multiple target areas according to the content information. For example, for a face image, the eyes can be used as a target area, the cheeks can be used as a target area, the mouth can be used as a target area, and so on. Then, a weight value is assigned to the target area according to the importance of the content and position in the target area, for example, the weight value of the eyes is higher and the weight value of the cheeks is lower.

[0162] Finally, the cloud computing node calculates the desensitization value of the desensitized erased area according to the target area covered by the desensitized erased area; if the desensitization value is not less than the preset threshold, the desensitized image is determined to be a desensitized qualified image; if the desensitization value is less than the preset threshold, the desensitized image is determined to be a desensitized unqualified image.

[0163] In the implementation mode of the present application, the desensitization value is obtained by adding the weight values ​​corresponding to the target area covered by the desensitized erasure area. When the desensitization value is not less than the preset threshold, it means that the important content in the sensitive information has been desensitized. Although it has not been completely erased, the specific information cannot be obtained. When the desensitization value is less than the preset threshold, it means that the important content in the sensitive information has not been desensitized, and it is still possible to cause information leakage and information insecurity, so the desensitized image is judged to be an unqualified desensitized image.

[0164] After the images are determined to be unqualified for desensitization, the number of abnormal images and the number of unqualified images are counted according to the specific unqualified situations. For example, the sensitive information is not desensitized, the sensitive information is not completely desensitized, etc.

[0165] By calculating the intersection and union ratio of the detection area and the desensitized erased area, the coverage of the desensitized erasure can be determined. For desensitized images whose intersection and union ratio of the detection area is not less than the preset intersection and union ratio, the desensitization value is no longer calculated, thereby reducing the test time and the use of computing resources. For desensitized images whose intersection and union ratio of the detection area is less than the preset intersection and union ratio, the desensitization value of the desensitized erased area is further calculated based on the content in the detection area and the content of the corresponding desensitized erased area, and the desensitization test result is determined based on the desensitization value, thereby improving the accuracy, completeness and reliability of the desensitization test on the test data set.

[0166] Step S2034: if any detection area does not include a desensitization erasure area, the desensitized image is determined to be a desensitization unqualified image;

[0167] If it is identified that a certain detection area in the desensitized image does not contain a desensitized erased area, it means that the detection area has not been desensitized, and the desensitized image can be directly determined as an unqualified desensitized image. For example, there are two detection areas, one of which contains address information and the other contains mobile phone number information. It is detected that only the detection area corresponding to the address information is desensitized, and the detection area corresponding to the mobile phone number information is not desensitized. In this case, the desensitized image is determined to be an unqualified image, and the abnormal situation of the unqualified image is that the detection area corresponding to the mobile phone number information is not desensitized.

[0168] Step S2035, determining the evaluation result of the data set to be tested based on the abnormal situation of the desensitization unqualified images in the data set to be tested.

[0169] In an implementation manner of the present application, the desensitization pass rate of the test data set of this image type is calculated based on the number of images in the test data set and the number of unqualified desensitized images. According to the abnormal conditions of the unqualified images, the abnormal rates are calculated for different abnormal conditions.

[0170] By detecting each detection area of ​​each desensitized image to perform desensitized erased area detection, when it is detected that there is no desensitized erased area in the detection area, the desensitized image can be directly determined as a desensitized unqualified image, thereby reducing the amount of calculation for cloud computing nodes to identify the contents in the detection area and the desensitized erased area, and reducing the test time and the use of computing resources; for desensitized images in which there is a desensitized erased area in each detection area, the desensitization test result is further determined based on the content in the detection area of ​​the desensitized image and the content in the corresponding desensitized erased area, thereby improving the reliability and accuracy of the desensitization test.

[0171] Step S204: Send the image desensitization test result to the control node device.

[0172] In the implementation manner of the present application, symmetric encryption, asymmetric encryption or other encryption is also used to send the image desensitization test result of the test data set to be tested to the control node device.

[0173] This method obtains desensitization test rules according to the image type of the data set to be tested through cloud computing nodes, which can more accurately evaluate the results of image desensitization testing and improve the accuracy of the test. This method can be used in conjunction with control node devices to process multiple data sets to be tested in a distributed manner and improve the overall processing capability. At the same time, this method supports expansion, and the testing capability can be improved by increasing the number of cloud computing nodes or improving the algorithm.

[0174] Figure 7 FIG. 7 is a structural block diagram of a control node device 700 according to an embodiment of the present application. Figure 7 As shown, the control node device 700 includes a processor 701 and a memory 702 , and may further include one or more of an information input / information output (I / O) interface 703 and a communication component 704 .

[0175] The processor 701 is used to control the overall operation of the control node device 700 to complete all or part of the steps in the above-mentioned image desensitization test method; the memory 702 is used to store various types of data to support the operation of the control node device 700, and these data may include, for example, instructions for any application or method used to operate on the control node device 700, and application-related data. The memory 702 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (Static Random Access Memory, SRAM), electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory, EEPROM), erasable programmable read-only memory (Erasable Programmable Read-Only Memory, EPROM), programmable read-only memory (Programmable Read-Only Memory, PROM), read-only memory (Read-Only Memory, ROM), magnetic memory, flash memory, disk or optical disk. One or more.

[0176] The I / O interface 703 provides an interface between the processor 701 and other interface modules, and the above-mentioned other interface modules can be keyboards, mice, buttons, etc. These buttons can be virtual buttons or physical buttons. The communication component 704 is used to test the wired or wireless communication between the control node device 700 and other devices. Wireless communication, such as Wi-Fi, Bluetooth, Near Field Communication (NFC), 2G, 3G or 4G, or a combination of one or more of them, so the corresponding communication component 704 may include: Wi-Fi components, Bluetooth components, NFC components.

[0177] The communication bus 705 may include a path to transmit information between the above components. The communication bus 705 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc. The communication bus 705 may be divided into an address bus, a data bus, a control bus, etc.

[0178] The control node device 300 can be implemented by one or more application specific integrated circuits (ASIC), digital signal processors (DSP), digital signal processing devices (DSPD), programmable logic devices (PLD), field programmable gate arrays (FPGA), controllers, microcontrollers, microprocessors or other electronic components to execute the image desensitization test method given in the above embodiment.

[0179] The computer-readable storage medium provided in an embodiment of the present application is introduced below. The computer-readable storage medium described below and the image desensitization testing method described above can be referenced to each other.

[0180] The present application also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the above-mentioned image desensitization testing method are implemented.

[0181] The computer-readable storage medium may include: a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and other media that can store program codes.

[0182] The terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not expressly listed, or also includes elements inherent to such process, method, article, or apparatus.

[0183] The above description is only a preferred embodiment of the present application and an explanation of the technical principles used. Those skilled in the art should understand that the scope of application involved in the present application is not limited to the technical solution formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the aforementioned application concept. For example, the above features are replaced with (but not limited to) technical features with similar functions applied in the present application.

Claims

1. An image desensitization testing method, characterized in that: include: Acquire multiple data sets to be tested; wherein the image data contained in the same data set to be tested have the same image type; According to the data size and image type of each data set to be tested, determine the final computing node of each data set to be tested from multiple cloud computing nodes, so that the final computing node performs an image desensitization test on the assigned data set to be tested, and generates an image desensitization test result for each data set to be tested; The number of the cloud computing nodes is not less than the number of the data sets to be tested; and determining the final computing node of each data set to be tested from the plurality of cloud computing nodes according to the data size and image type of each data set to be tested comprises: Obtaining desensitization type information of each of the cloud computing nodes; wherein the desensitization type information includes at least one desensitization type; and the desensitization type corresponds to the image type one by one; Matching at least one to-be-allocated computing node for each of the data sets to be tested based on the image type of the data sets to be tested and the desensitization type information; Determine whether there is a test data set that matches only one computing node to be assigned; If there is a data set to be tested that matches only one computing node to be allocated, the computing node to be allocated that matches the data set to be tested is determined as the final computing node of the data set to be tested, and the final computing node is recorded as the allocated computing node, and the step of determining whether there is a data set to be tested that matches only one computing node to be allocated is performed until there is no data set to be tested that matches only one computing node to be allocated; If there is no data set to be tested that matches only one computing node to be allocated, a data set to be tested that matches multiple computing nodes to be allocated is determined as a target data set to be tested, and a final computing node of the target data set to be tested is determined according to the data size of the target data set to be tested and the computing resource information of the computing node to be allocated; The step of determining the final computing node of the target data set to be tested according to the data size of the target data set to be tested and the computing resource information of the computing node to be allocated comprises: Sorting all target test data sets according to the data size of the test data sets to obtain a first sorting result; Obtain computing resource information of each computing node to be allocated; wherein the computing resource information includes memory, processing speed and network bandwidth; Calculate a score value of each to-be-allocated computing node according to the computing resource information; Specifically, the memory, processing speed and network bandwidth are weighted, and the memory usage, processing speed ratio and network bandwidth ratio are calculated. Then, the weighted sum is taken to obtain the score value of the computing node to be allocated. According to the score value, sorting the multiple to-be-allocated computing nodes corresponding to each of the target data sets to be tested, to obtain a second sorting result corresponding to the target data sets to be tested; Select target data sets to be tested in sequence according to the first sorting result; for the selected target data sets to be tested, determine the to-be-allocated computing node with the highest score value in the second sorting result corresponding to the selected target data sets to be tested as the final computing node according to the second sorting result corresponding to the selected target data sets to be tested, and record the final computing node as the allocated computing node; until the final computing node of each target data set to be tested is determined; after determining the final computing node of each data set to be tested from multiple cloud computing nodes according to the data size and image type of each data set to be tested, the method further includes: Obtaining a preset time threshold for each data set to be tested based on the data size of the data set to be tested; The preset time threshold is a value preset in advance based on historical execution conditions and the data size of the data set to be tested; Based on the processing speed, the network bandwidth and the data size of the data set to be tested, predict the test time of each of the cloud computing nodes processing the assigned data set to be tested; Determine whether there is a cloud computing node whose test time is greater than the preset time threshold; If there is a computing node whose test time is greater than the preset time threshold, the to-be-tested data set corresponding to the cloud computing node whose test time is greater than the preset time threshold is used as the to-be-divided data set, and the cloud computing node whose test time is greater than the preset time threshold is used as the first node; Determine whether there is a cloud computing node that processes the image type of the data set to be divided and the test time is less than the preset time threshold; If there is a cloud computing node that tests the image type of the data set to be divided and the test time is less than the preset time threshold, the cloud computing node is used as the second node; the data set to be divided is divided based on the test time of the first node and the preset time threshold to obtain a first divided data set and a second divided data set; Determining whether the second node can process the second divided data set within the preset time threshold; If the second node can process the second divided data set within the preset time threshold, the second divided data set is allocated to the second node.

2. The method according to any one of claim 1, characterized in that: The matching of at least one to-be-allocated computing node for each of the data sets to be tested based on the image type of the data set to be tested and the desensitization type information comprises: For any of the data sets to be tested, a cloud computing node corresponding to a desensitization type that corresponds one-to-one to the image type of the data set to be tested is searched from the multiple cloud computing nodes, and the found cloud computing node is used as a computing node to be allocated for the data set to be tested.

3. The method according to claim 1, characterized in that Also includes: Obtaining the image desensitization test result returned by each of the final computing nodes; The desensitization test results of each image type are analyzed based on the image desensitization test results.

4. An image desensitization testing device, characterized in that: include: A first acquisition module is used to acquire multiple data sets to be tested; wherein the image data contained in the same data set to be tested have the same image type; A first determination module is used to determine the final computing node of each of the data sets to be tested from multiple cloud computing nodes according to the data size and image type of each of the data sets to be tested, so that the final computing node performs an image desensitization test on the assigned data set to be tested, and generates an image desensitization test result for each of the data sets to be tested; Wherein, the number of the cloud computing nodes is not less than the number of the data sets to be tested; An acquisition submodule is used to acquire desensitization type information of each cloud computing node; wherein the desensitization type information includes at least one desensitization type; and the desensitization type corresponds to the image type one by one; A matching submodule, used for matching at least one to-be-allocated computing node for each data set to be tested based on the image type and desensitization type information of the data set to be tested; A judgment submodule is used to judge whether there is a test data set that matches only one computing node to be assigned; A first determination submodule is used for, if there is a data set to be tested that matches only one computing node to be allocated, determining the computing node to be allocated that matches the data set to be tested as the final computing node of the data set to be tested, and recording the final computing node as the allocated computing node, and executing the step of determining whether there is a data set to be tested that matches only one computing node to be allocated, until there is no data set to be tested that matches only one computing node to be allocated; The second determination submodule is used to determine the test data set matching multiple computing nodes to be assigned as the target test data set if there is no test data set matching only one computing node to be assigned, and determine the final computing node of the target test data set according to the data size of the target test data set and the computing resource information of the computing node to be assigned; The second determination submodule is specifically used to sort all target data sets to be tested according to the data size of the data sets to be tested to obtain a first sorting result; obtain computing resource information of each computing node to be allocated; wherein the computing resource information includes memory, processing speed and network bandwidth; calculate the score value of each computing node to be allocated according to the computing resource information; specifically, set respective weights for memory, processing speed and network bandwidth, and calculate the memory occupancy, processing speed ratio and network bandwidth ratio, and then perform weighted summation to obtain the score value of the computing node to be allocated; sort the multiple computing nodes to be allocated corresponding to each target data set to be tested according to the score value to obtain a second sorting result corresponding to the target data set to be tested; select the target data set to be tested in turn according to the first sorting result; for the selected target data set to be tested, according to the second sorting result corresponding to the selected target data set to be tested, determine the computing node to be allocated with the highest score value in the second sorting result corresponding to the selected target data set to be tested as the final computing node, and record the final computing node as the allocated computing node; until the final computing node of each target data set to be tested is determined; A second acquisition module is used to obtain a preset time threshold of each data set to be tested based on the data size of the data set to be tested, after respectively determining the final computing node of each data set to be tested from multiple cloud computing nodes according to the data size and image type of each data set to be tested; wherein the preset time threshold is a value preset in advance according to the historical execution status and the data size of the data set to be tested; A prediction module, used for predicting the test time of each cloud computing node processing the assigned data set to be tested based on the processing speed, network bandwidth and the data size of the data set to be tested; A first judgment module is used to judge whether there is a cloud computing node whose test time is greater than a preset time threshold; A second determination module is used to, if there is a computing node whose test time is greater than a preset time threshold, use the to-be-tested data set corresponding to the cloud computing node whose test time is greater than the preset time threshold as the to-be-divided data set, and use the cloud computing node whose test time is greater than the preset time threshold as the first node; The second judgment module is used to judge whether there is a cloud computing node that processes the image type of the data set to be divided and the test time is less than a preset time threshold; A partitioning module is determined, which is used to use the cloud computing node as the second node if there is a cloud computing node that tests the image type of the data set to be partitioned and the test time is less than a preset time threshold; the data set to be partitioned is partitioned based on the test time of the first node and the preset time threshold to obtain a first partitioned data set and a second partitioned data set; A third judgment module is used to judge whether the second node can process the second divided data set within a preset time threshold; The allocation module is configured to allocate the second divided data set to the second node if the second node can process the second divided data set within a preset time threshold.

5. A control node device, characterized in that: comprising a processor coupled to a memory; The processor is used to execute the computer program stored in the memory, so that the control node device executes the method according to any one of claims 1 to 3.

6. A cloud service platform, characterized in that: It comprises the control node device as claimed in claim 5, a cloud storage node, a cloud network node and multiple cloud computing nodes.

7. A computer-readable storage medium, characterized in that: The method comprises a computer program or an instruction, and when the computer program or the instruction is executed on a computer, the computer is caused to execute the method according to any one of claims 1 to 3.

Images