Blockchain-based internet image big data intelligent storage method and system

Abstract

The application discloses an internet image big data intelligent storage method and system based on a block chain, and relates to the technical field of data analysis.The method comprises the following steps: recognizing image text information, and deleting the corresponding image if the recognition result contains preset illegal words; otherwise, describing the image, and deleting the corresponding image if the description result contains preset illegal words; otherwise, carrying out noise detection on the core area of the image, and deleting the corresponding image if the detection result is less than a preset signal-to-noise ratio threshold; otherwise, carrying out semantic segmentation on the image, normally storing or storing after compressing the image according to the semantic segmentation result, and uploading the network staff identity, compression time and compression result to a chain.The application utilizes a plurality of model algorithms to cooperate with each other to accurately recognize and analyze the image, and then reasonably store the image; meanwhile, important information is uploaded to a chain in combination with the block chain technology, so that the data safety is ensured.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data analysis, in particular to an internet image big data intelligent storage method and system based on a blockchain. BACKGROUND

[0002] With the vigorous development of the Internet, more and more users upload images to the Internet. However, if part of the illegal images, high noise images cannot be deleted in time, not only will affect the user's online experience, but also will occupy a large storage resource. At the same time, the high-precision storage of non-semantic rich images will also occupy more storage resources. Therefore, it is necessary to analyze the image in depth, determine whether to retain the image, and determine whether the retained image needs to be compressed and then stored reasonably, so as to save storage resources. However, in the image compression process, the problem of malicious tampering may be involved, so the security of the image compression process also needs to be controlled, and the blockchain technology can provide direct support for data security. Therefore, how to realize the intelligent storage of internet image big data by combining the blockchain technology has become a problem to be solved. SUMMARY

[0003] In order to overcome the above problems or at least partially solve the above problems, the embodiments of the present application provide an internet image big data intelligent storage method and system based on a blockchain, which uses a variety of model algorithms such as an OCR model generated based on multi-level noise, an image description model based on multi-scale mutual inspection, a peak signal-to-noise ratio detection model based on salient region, and a semantic segmentation model based on sharpening convolution kernel filtering to cooperate with each other, to accurately identify and analyze the image, and then to store it reasonably. At the same time, important information is combined with the blockchain technology to be chained, so as to ensure data security.

[0004] The embodiments of the present application are implemented as follows:

[0005] In a first aspect, the embodiments of the present application provide an internet image big data intelligent storage method based on a blockchain, comprising the following steps:

[0006] For any image in the Internet, an OCR model based on multi-level noise generation is used to identify the image text information to obtain an identification result;

[0007] If the identification result contains a preset illegal vocabulary, the corresponding image is deleted; otherwise, an image description model based on multi-scale mutual inspection is used to describe the image to generate a description result;

[0008] If the description result contains a preset illegal vocabulary, the corresponding image is deleted; otherwise, a peak signal-to-noise ratio detection model based on salient region is used to detect the noise of the core region of the image to generate a detection result;

[0009] If the detection result is less than the preset signal-to-noise ratio threshold, the corresponding image is deleted; otherwise, a semantic segmentation model based on sharpening convolution kernel filtering is used to perform semantic segmentation on the image to obtain a semantic segmentation result.

[0010] According to the semantic segmentation result, the image is normally stored or sent to a network worker for compression processing, the compressed image is stored, and the network worker identity information, image compression processing time and image compression result are uploaded to the blockchain.

[0011] To solve the problems in the prior art, the method uses an OCR model based on multi-level noise generation to accurately identify the image text information, providing direct support for effectively deleting images containing illegal words. Based on the text recognition, an image description model based on multi-scale mutual verification is used to accurately describe the image, providing direct support for effectively deleting images containing illegal semantic content. Then, a peak signal-to-noise ratio detection model based on salient region is used to accurately detect whether the core region of the image contains significant noise, providing direct support for deleting images with significant noise in the core region. On this basis, the method also uses a semantic segmentation model based on sharpening convolution kernel filtering to accurately segment the image, providing direct support for determining whether the image is a semantic-rich image. For images that need to be compressed and stored, the blockchain technology is used to chain store the important information such as the identity of the network worker, the time of compressing the image, and the image compression result, effectively preventing malicious modification during the image compression process and improving the reliability and security of the system. The present invention combines the OCR model based on multi-level noise generation, the image description model based on multi-scale mutual verification, the peak signal-to-noise ratio detection model based on salient region, and the semantic segmentation model based on sharpening convolution kernel filtering, etc. to accurately identify and analyze the image, and then store it reasonably. At the same time, the blockchain technology is combined to chain important information, ensuring data security.

[0012] Based on the first aspect, in some embodiments of the present invention, the method of identifying the image text information using the OCR model based on multi-level noise generation to generate an identification result includes the following steps:

[0013] Add multiple levels of noise to the image to obtain multiple new images;

[0014] Use the OCR model to identify the text information of each new image to obtain and determine the final identification result according to the corresponding multiple text identification results.

[0015] In some embodiments of the first aspect, the method for describing an image and generating a description result by using the image description model based on multi-scale mutual verification comprises the following steps:

[0016] The image is divided into multiple scales to obtain images of multiple scales.

[0017] Each image of each scale is described to obtain a final description result according to multiple scale description results.

[0018] In some embodiments of the first aspect, the method for detecting noise in a core region of an image and generating a detection result by using the peak signal-to-noise ratio detection model based on a salient region comprises the following steps:

[0019] The salient region of the image is detected to obtain a salient region image.

[0020] The peak signal-to-noise ratio of the salient region image is calculated to generate a detection result.

[0021] In some embodiments of the first aspect, the method for performing semantic segmentation on an image and obtaining a semantic segmentation result by using the semantic segmentation model based on a sharpening convolution kernel filter comprises the following steps:

[0022] The image is filtered by using a sharpening convolution kernel to obtain a filtered image.

[0023] The filtered image is subjected to semantic segmentation to obtain multiple semantic segmentation region images.

[0024] In the second aspect, the embodiments of the present application provide an internet image big data intelligent storage system based on a blockchain, which comprises a character recognition module, an image description module, a signal-to-noise ratio detection module, a semantic segmentation module, and a storage module.

[0025] The character recognition module is configured to recognize the character information of an image in the internet by using an OCR model based on multi-level noise generation to obtain a recognition result.

[0026] The image description module is configured to delete the corresponding image if the recognition result contains preset illegal words, and otherwise, describe the image by using an image description model based on multi-scale mutual verification to generate a description result.

[0027] The signal-to-noise ratio detection module is configured to delete the corresponding image if the description result contains preset illegal words, and otherwise, detect noise in a core region of the image by using a peak signal-to-noise ratio detection model based on a salient region to generate a detection result.

[0028] The semantic segmentation module is configured to: if the detection result is less than a preset signal-to-noise ratio threshold, delete the corresponding image; otherwise, perform semantic segmentation on the image by using a semantic segmentation model based on a sharpening convolution kernel filtering to obtain a semantic segmentation result.

[0029] The storage module is configured to: store the image normally according to the semantic segmentation result, or send the corresponding image to a network staff for compression processing, store the image after compression processing, and upload network staff identity information, image compression processing time and image compression result to a blockchain.

[0030] To solve the problems in the prior art, the system cooperates a plurality of modules such as the character recognition module, the image description module, the signal-to-noise ratio detection module, the semantic segmentation module and the storage module, uses an OCR model based on multi-level noise generation to accurately recognize image character information, provides direct support for effectively deleting images containing illegal words, on the basis of character recognition, uses an image description model based on multi-scale mutual verification to accurately describe the image, provides direct support for effectively deleting images containing illegal semantic content, then uses a peak signal-to-noise ratio detection model based on salient region to accurately detect whether the core region of the image contains significant noise, provides direct support for deleting images with significant noise in the core region, on this basis, the system also uses a semantic segmentation model based on sharpening convolution kernel filtering to accurately segment the image, provides direct support for judging whether the image is a semantic-rich image, for images that need to be compressed and stored, the system uses blockchain technology to respectively chain store important information such as network staff identity, image compression time and image compression result, effectively prevents malicious modification during image compression, and improves the reliability and security of the system. The present application combines the OCR model based on multi-level noise generation, the image description model based on multi-scale mutual verification, the peak signal-to-noise ratio detection model based on salient region and the semantic segmentation model based on sharpening convolution kernel filtering, accurately recognizes and analyzes the image, and then reasonably stores the image, and chain stores important information by using the blockchain technology, thereby ensuring data security.

[0031] In a third aspect, an electronic device is provided. The electronic device includes a memory configured to store one or more programs; and a processor. When the one or more programs are executed by the processor, the method of any one of the above first aspect is implemented.

[0032] In a fourth aspect, a computer-readable storage medium is provided. The computer-readable storage medium stores a computer program. When the computer program is executed by a processor, the method of any one of the above first aspect is implemented.

[0033] The embodiments of the present application have at least the following advantages or beneficial effects:

[0034] The embodiment of the present application provides an internet image big data intelligent storage method and system based on a blockchain, which combines various model algorithms such as an OCR model generated based on multi-level noise, an image description model based on multi-scale mutual inspection, a peak signal-to-noise ratio detection model based on a salient region, and a semantic segmentation model based on a sharpening convolution kernel filter, to accurately identify and analyze images, and then store the images reasonably; meanwhile, important information is uploaded to the blockchain by combining the blockchain technology, so that the data security is ensured. BRIEF DESCRIPTION OF DRAWINGS

[0035] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiments. It should be understood that the following drawings only show some embodiments of the present application, and therefore should not be regarded as a limitation to the scope. For those skilled in the art, other related drawings can also be obtained without creative labor on the basis of these drawings.

[0036] Figure 1 A flowchart of an internet image big data intelligent storage method based on a blockchain according to an embodiment of the present application;

[0037] Figure 2 A flowchart of identifying image text information by using an OCR model generated based on multi-level noise in an internet image big data intelligent storage method based on a blockchain according to an embodiment of the present application;

[0038] Figure 3 A flowchart of describing an image by using an image description model based on multi-scale mutual inspection in an internet image big data intelligent storage method based on a blockchain according to an embodiment of the present application;

[0039] Figure 4 A principle block diagram of an internet image big data intelligent storage system based on a blockchain according to an embodiment of the present application;

[0040] Figure 5 A structure block diagram of an electronic device according to an embodiment of the present application is provided.

[0041] Explanation of reference signs: 100, text recognition module; 200, image description module; 300, signal-to-noise ratio detection module; 400, semantic segmentation module; 500, storage module; 101, storage; 102, processor; 103, communication interface. DETAILED DESCRIPTION

[0042] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0043] Therefore, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention.

[0044] It should be noted that similar labels and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures.

[0045] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0046] In the description of the embodiments of the present invention, "multiple" means at least two.

[0047] Example:

[0048] like Figures 1-3 As shown, in a first aspect, embodiments of the present invention provide a blockchain-based intelligent storage method for internet image big data, comprising the following steps:

[0049] S1. For any image on the Internet, use an OCR model based on multi-level noise generation to recognize the text information in the image to obtain the recognition result;

[0050] Furthermore, such as Figure 2 As shown, it includes:

[0051] S11. Add multiple levels of noise to the image to obtain multiple new images;

[0052] S12. Use an OCR model to recognize the text information of each new image to obtain and determine the final recognition result based on the corresponding multiple text recognition results.

[0053] In some embodiments of the present invention, for any image on the Internet, the text information in the image is identified using an OCR model based on multi-level noise generation. If the image is determined to contain illegal words, the image is directly deleted; otherwise, the next step is performed.

[0054] The method described above for recognition based on an OCR model generated from multi-level noise includes: adding different levels of noise (usually three levels are sufficient, but high noise levels should be avoided) to the image to obtain multiple new images, and then using OCR technology to recognize each new image. If the recognition results of the vast majority of new images contain illegal words, the text information of the image is deemed to contain illegal words, and the image is directly deleted.

[0055] S2. If the recognition result contains preset illegal words, the corresponding image will be deleted; otherwise, the image description model based on multi-scale cross-verification will be used to describe the image and generate a description result.

[0056] Furthermore, such as Figure 3 As shown, it includes:

[0057] S21. Divide the image into multiple scales to obtain images of multiple scales;

[0058] S22. Describe the images at each scale separately to obtain and determine the final description result based on the description results at multiple scales.

[0059] In some embodiments of the present invention, an image description model based on multi-scale cross-verification is used to describe the image. If the image is determined to contain illegal semantic content, the image is directly deleted; otherwise, the next step is performed.

[0060] The aforementioned image description model based on multi-scale cross-verification describes an image by: decomposing the image into images of multiple scales, and describing each scale separately. If most of the description results contain illegal words, the image is deemed to contain illegal semantic content and is directly deleted.

[0061] S3. If the description result contains preset illegal words, the corresponding image will be deleted; otherwise, the peak signal-to-noise ratio detection model based on salient regions will be used to detect noise in the core region of the image and generate the detection result.

[0062] Further, the method comprises: performing salient region detection on the image to obtain a salient region image; calculating a peak signal-to-noise ratio of the salient region image to generate a detection result.

[0063] In some embodiments of the present application, a salient region-based peak signal-to-noise ratio detection model is used to detect noise in a core region of the image. If the noise in the core region of the image is significant, the image is directly deleted; otherwise, the next step is performed.

[0064] The salient region-based peak signal-to-noise ratio detection model described above for detecting noise in a core region of an image comprises: performing salient region detection on the image and calculating a peak signal-to-noise ratio of the salient region. If the peak signal-to-noise ratio of the salient region is low, lower than a preset signal-to-noise ratio threshold, it is determined that the noise in the core region of the image is significant, and the image is directly deleted.

[0065] S4, if the detection result is less than the preset signal-to-noise ratio threshold, the corresponding image is deleted; otherwise, a sharpening convolution kernel-based filtering semantic segmentation model is used to perform semantic segmentation on the image to obtain a semantic segmentation result.

[0066] Further, the method comprises: filtering the image based on the sharpening convolution kernel to obtain a filtered image; and performing semantic segmentation on the filtered image to obtain a plurality of semantic segmentation region images.

[0067] S5, according to the semantic segmentation result, the image is normally stored or sent to a network worker for compression processing, the compressed image is stored, and the network worker's identity information, image compression processing time and image compression result are uploaded to the blockchain.

[0068] In some embodiments of the present application, a sharpening convolution kernel-based filtering semantic segmentation model is used to obtain a semantic segmentation result of the image. If the image is segmented into more semantic regions, it is determined that the image is a semantic-rich image, and the image is normally stored. If the image is segmented into fewer semantic regions, it is determined that the image is a non-semantic-rich image, and a network worker compresses the image (mainly using an image compression method that can still be easily visualized after compression), and then stores the compressed image. The network worker's identity, the time of compressing the image, and the image compression result are respectively stored on the chain.

[0069] For all images, the above method is used for analysis, and whether to retain the image is determined according to the analysis result. For the retained image, if it is determined as a semantic-rich image, the image is normally stored; if it is determined as a non-semantic-rich image, the network staff compresses the image, and then stores the compressed image. At the same time, important information in the compression process is stored in the chain.

[0070] In order to solve the problems in the prior art, the method uses an OCR model based on multi-level noise generation to accurately identify the image text information, thereby providing direct support for effectively deleting images containing illegal words. On the basis of text recognition, an image description model based on multi-scale mutual verification is used to accurately describe the image, thereby providing direct support for effectively deleting images containing illegal semantic content. Then, a peak signal-to-noise ratio detection model based on salient region is used to accurately detect whether the core region of the image contains significant noise, thereby providing direct support for deleting images with significant noise in the core region. On this basis, the method also uses a semantic segmentation model based on sharpening convolution kernel filtering to accurately segment the image, thereby providing direct support for determining whether the image is a semantic-rich image. For images that need to be compressed and stored, the blockchain technology is used to store the important information such as the identity of the network staff, the time of compressing the image, and the image compression result in the chain, thereby effectively preventing malicious modification during the image compression process and improving the reliability and security of the system. The present application combines the OCR model based on multi-level noise generation, the image description model based on multi-scale mutual verification, the peak signal-to-noise ratio detection model based on salient region, and the semantic segmentation model based on sharpening convolution kernel filtering, and other model algorithms to accurately identify and analyze the image, and then store it reasonably. At the same time, the blockchain technology is combined to store important information in the chain, thereby ensuring data security.

[0071] As shown in Figure 4 the second aspect, the embodiments of the present application provide an internet image big data intelligent storage system based on blockchain, which comprises a text recognition module 100, an image description module 200, a signal-to-noise ratio detection module 300, a semantic segmentation module 400, and a storage module 500, wherein:

[0072] The text recognition module 100 is used to identify the image text information of any image in the internet by using an OCR model based on multi-level noise generation to obtain an identification result.

[0073] The image description module 200 is used to delete the corresponding image if the identification result contains preset illegal words; otherwise, an image description model based on multi-scale mutual verification is used to describe the image to generate a description result.

[0074] The signal-to-noise ratio detection module 300 is configured to: if the description result contains preset illegal words, the corresponding image is deleted; otherwise, the core region of the image is subjected to noise detection by using a peak signal-to-noise ratio detection model based on a salient region to generate a detection result.

[0075] The semantic segmentation module 400 is configured to: if the detection result is less than a preset signal-to-noise ratio threshold, the corresponding image is deleted; otherwise, the image is subjected to semantic segmentation by using a semantic segmentation model based on a sharpening convolution kernel filtering to obtain a semantic segmentation result.

[0076] The storage module 500 is configured to: according to the semantic segmentation result, the image is normally stored or the corresponding image is sent to a network staff for compression processing, the image subjected to the compression processing is stored, and the network staff identity information, the image compression processing time and the image compression result are uploaded to a blockchain.

[0077] To solve the problems in the prior art, the system cooperates the text recognition module 100, the image description module 200, the signal-to-noise ratio detection module 300, the semantic segmentation module 400 and the storage module 500, uses an OCR model based on multi-level noise generation to accurately recognize image text information, provides direct support for effectively deleting images containing illegal words. On the basis of text recognition, an image description model based on multi-scale mutual verification is used to accurately describe the image, which provides direct support for effectively deleting images containing illegal semantic content. Then, a peak signal-to-noise ratio detection model based on salient region is used to accurately detect whether the core region of the image contains significant noise, which provides direct support for deleting images with significant noise in the core region. On this basis, the system also uses a semantic segmentation model based on sharpening convolution kernel filtering to accurately segment the image, which provides direct support for judging whether the image is a semantic-rich image. For images that need to be compressed and stored, the blockchain technology is used to chain store the important information such as the network staff identity, the image compression time and the image compression result, effectively preventing the image compression process from being maliciously modified, and improving the reliability and security of the system. The present application combines the OCR model based on multi-level noise generation, the image description model based on multi-scale mutual verification, the peak signal-to-noise ratio detection model based on salient region and the semantic segmentation model based on sharpening convolution kernel filtering, accurately recognizes and analyzes the image, and then reasonably stores the image. At the same time, the important information is chained by using the blockchain technology, so as to ensure data security.

[0078] As Figure 5In the third aspect, the embodiments of the present application provide an electronic device, which includes a memory 101 configured to store one or more programs; and a processor 102. When the one or more programs are executed by the processor 102, the method according to any one of the first aspect is implemented.

[0079] The electronic device further includes a communication interface 103. The memory 101, the processor 102 and the communication interface 103 are electrically connected with each other directly or indirectly to achieve the transmission or interaction of data. For example, the elements can be electrically connected with each other through one or more communication buses or signal lines. The memory 101 can be configured to store software programs and modules. The processor 102 can execute various functional applications and data processing by executing the software programs and modules stored in the memory 101. The communication interface 103 can be configured to perform signaling or data communication with other node devices.

[0080] The memory 101 can be, but is not limited to, a random access memory (RAM), a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM) and the like.

[0081] The processor 102 can be an integrated circuit chip with a signal processing capability. The processor 102 can be a general purpose processor, including a central processing unit (CPU), a network processor (NP) and the like; or can be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

[0082] In the embodiments of the present application, it should be understood that the disclosed method and system and method can also be implemented by other manners. The above-described method and system embodiments are only illustrative, for example, the flowcharts and block diagrams in the drawings show the possible implementation architecture, function and operation of the method and system, method and computer program product according to the embodiments of the present application. In this regard, each block in the flowchart or block diagram can represent a module, a program segment or a part of code, which includes one or more executable instructions for implementing the specified logic function. It should also be noted that in some alternative implementations, the functions noted in the blocks can occur in different order from that shown in the drawings. For example, two consecutive blocks can actually be executed substantially in parallel, and sometimes they can be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagram and / or flowchart, and the combination of blocks in the block diagram and / or flowchart, can be implemented by a dedicated hardware-based system for performing the specified functions or actions, or can be implemented by a combination of dedicated hardware and computer instructions.

[0083] In addition, the functional modules in the various embodiments of the present application can be integrated together to form an independent part, or each module can exist independently, or two or more modules can be integrated to form an independent part.

[0084] In a fourth aspect, the embodiments of the present application provide a computer readable storage medium, which stores a computer program. The computer program is executed by the processor 102 to implement the method of any one of the above first aspect. The functions, if implemented in the form of software function modules and sold or used as independent products, can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in the various embodiments of the present application. The foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various media that can store program codes.

[0085] The above merely provides preferred embodiments of the present application, and is not intended to limit the present application. For those skilled in the present application, various modifications and variations can be made thereto. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present application shall fall within the scope of the present application.

[0086] It is apparent that the application is not limited to the details of the foregoing exemplary embodiments, and this application can be carried out in other specific forms without departing from the spirit or essential characteristics of the application. Accordingly, it is to be understood that the embodiments are to be considered in all respects as illustrative and not restrictive, and the scope of the application is to be governed by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are intended to be embraced therein. No reference signs in the claims shall be construed as limiting the scope of the claims.

Claims

1. A blockchain-based intelligent storage method for Internet image big data, characterized in that, The method comprises the following steps: For any image in the Internet, the OCR model based on multi-level noise generation is used to identify the image text information to obtain the recognition result; If the recognition result contains the preset illegal vocabulary, the corresponding image is deleted; otherwise, the image description model based on multi-scale mutual test is used to describe the image to generate a description result; If the description result contains the preset illegal vocabulary, the corresponding image is deleted; otherwise, the peak signal-to-noise ratio detection model based on the salient region is used to detect the noise of the core region of the image to generate a detection result; If the detection result is less than the preset signal-to-noise ratio threshold, the corresponding image is deleted; otherwise, the semantic segmentation model based on the sharpening convolution kernel filtering is used to perform semantic segmentation on the image to obtain a semantic segmentation result; According to the semantic segmentation result, the image is normally stored or sent to the network staff for compression processing, and the compressed image is stored, and the network staff's identity information, image compression processing time and image compression result are uploaded to the blockchain. 2.The blockchain-based Internet image big data intelligent storage method according to claim 1, characterized in that, The method for identifying the image text information by using the OCR model based on multi-level noise generation to generate the recognition result comprises the following steps: Add multiple levels of noise to the image to obtain multiple new images; The text information of each new image is identified by using the OCR model to obtain the final recognition result according to the corresponding multiple text recognition results. 3.The blockchain-based internet image big data intelligent storage method of claim 1, wherein, The method for describing the image by using the image description model based on multi-scale mutual test to generate the description result comprises the following steps: The image is divided into multiple scales to obtain multiple scale images; Each scale image is described to obtain the final description result according to the multiple scale description results. 4.The blockchain-based Internet image big data intelligent storage method of claim 1, wherein, The method for detecting the noise of the core region of the image by using the peak signal-to-noise ratio detection model based on the salient region to generate the detection result comprises the following steps: The salient region of the image is detected to obtain a salient region image; The peak signal-to-noise ratio of the salient region image is calculated to generate a detection result. 5.The blockchain-based internet image big data intelligent storage method of claim 1, wherein, The method for performing semantic segmentation on the image by using the semantic segmentation model based on the sharpening convolution kernel filtering to obtain the semantic segmentation result comprises the following steps: The image is filtered based on the sharpening convolution kernel to obtain a filtered image; The filtered image is subjected to semantic segmentation to obtain multiple semantic segmentation region images.

6. A blockchain-based Internet image big data intelligent storage system, characterized in that, It comprises: a text recognition module, an image description module, a signal-to-noise ratio detection module, a semantic segmentation module and a storage module, wherein: The text recognition module is used to identify the image text information by using the OCR model based on multi-level noise generation for any image in the Internet to obtain the recognition result; The image description module is used to delete the corresponding image if the recognition result contains the preset illegal vocabulary; otherwise, the image description model based on multi-scale mutual test is used to describe the image to generate a description result; The signal-to-noise ratio detection module is configured to delete the corresponding image if the description result contains the preset illegal word; otherwise, the core region of the image is subjected to noise detection by using a peak signal-to-noise ratio detection model based on a salient region to generate a detection result. The semantic segmentation module is configured to delete the corresponding image if the detection result is less than a preset signal-to-noise ratio threshold; otherwise, the image is subjected to semantic segmentation by using a semantic segmentation model based on a sharpening convolution kernel filter to obtain a semantic segmentation result. The storage module is configured to normally store the image according to the semantic segmentation result or send the corresponding image to a network staff for compression processing of the image, store the image after compression processing, and upload network staff identity information, image compression processing time and image compression result to a blockchain.

7. An electronic device, comprising: comprise: a memory storing one or more programs; a processor; when the one or more programs are executed by the processor, the method of any one of claims 1-5 is implemented.

8. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to implement the method of any one of claims 1-5.

Images