Cross-system data co-processing method and system based on image recognition and chip storage

Abstract

The present application relates to a cross-system data co-processing method and system based on image recognition and chip storage. The method comprises the following steps: acquiring an electronic image of physical medium data by means of an image collection device or a virtual printer, pre-processing the electronic image, and using an image recognition algorithm to perform data positioning and extraction on the pre-processed image; writing the extracted data into a specified sector of an embedded chip, and verifying data consistency by means of a CRC32 check and read-back comparison mechanism; on the basis of encoding requirements of a target system, performing context-aware format adaptation on chip data by means of a dynamic format conversion model, and using an API interface and a SendKeys method to perform cross-system automated data transmission; and generating a whole-process operation log, wherein the operation log includes a timestamp, a chip serial number, a data hash value and a blockchain tamper-proof storage identifier, and supports abnormal backtracking and auditing.

Description

A cross-system data collaborative processing method and system based on image recognition and chip storage Technical Field

[0001] This application relates to the fields of data management and automation technology, and more specifically, to a cross-system data collaborative processing method and system based on image recognition and chip storage. Background Technology

[0002] Current cross-system data collaborative processing solutions generally suffer from the following technical bottlenecks: First, the digitization of physical media data (such as paper documents and barcode labels) relies on traditional image processing techniques. Existing methods (such as fixed-parameter Gaussian filtering and standard Otsu segmentation) are poorly adaptable to distorted and low-contrast images, resulting in a high text positioning error rate. Second, data storage verification mechanisms are simplistic; most solutions only use CRC or MD5 checksums, which cannot detect physical bit flip errors in the storage medium and lack cross-verification logic between image features and chip data. Third, format conversion during cross-system transmission relies on manually configured templates, making it difficult to dynamically adapt to the encoding differences of heterogeneous systems. Operation logs are mostly stored in centralized databases, posing a risk of tampering, while existing blockchain solutions suffer from storage latency exceeding 500ms in high-concurrency scenarios due to their fixed block capacity design. These problems severely restrict the demand for high-precision, real-time data collaboration in fields such as healthcare and logistics.

[0003] Existing technologies, such as the Chinese patent application with publication number "CN112100149A", disclose an automated log analysis system. This invention is achieved through the following technical solution: a log preprocessing module sends the cleaned work log file to a text analysis slicing module for semantic analysis, calculates the Simhash fingerprint of adjacent text in the log, and performs similarity calculation and judgment to form the minimum non-repeating text block; a parameter name and value extraction module constructs a regular expression matching pattern based on the minimum non-repeating text block or a pre-written log template, and selects the read log for parameter name and value extraction processing; a data IO module saves the extracted key text information, parameter names, and values ​​to the hard disk for subsequent reading and writing; and a statistical plotting and report generation module performs data statistics by category on the values ​​extracted from the hard disk, draws visualization graphs based on the statistical values, and generates an analysis report.

[0004] The problems with the aforementioned existing technologies are that this method only supports text log analysis and cannot process physical media data such as images and chip storage; it relies on conventional storage (such as CSV / database) and lacks the ability to verify physical media data; format conversion relies on predefined templates and cannot dynamically adapt to heterogeneous systems; and log storage relies on traditional databases, which poses a risk of tampering. Summary of the Invention

[0005] To solve the above technical problems, the application provides a cross-system data collaborative processing method and system based on image recognition and chip storage.

[0006] The technical scheme of the application is as follows:

[0007] The application provides a cross-system data collaborative processing method based on image recognition and chip storage, which comprises the following steps:

[0008] An electronic image of physical medium data is acquired by an image acquisition device or a virtual printer, and the electronic image is preprocessed, and an image recognition algorithm is used to position and extract data from the preprocessed image;

[0009] The extracted data is written into a specified sector of an embedded chip, and the data consistency is verified through a CRC32 checksum and read-back comparison mechanism;

[0010] According to the encoding requirements of the target system, a dynamic format conversion model is used for context-aware format adaptation of the chip data, and an API interface and a SendKeys method are used for automatic data transmission across systems;

[0011] A full-process operation log is generated, the operation log contains a timestamp, a chip serial number, a data hash value and a blockchain tamper-proof storage identifier, and supports abnormal backtracking and auditing.

[0012] As a preferred embodiment, the preprocessed electronic image specifically includes noise reduction filtering, geometric correction and adaptive threshold segmentation based on improved Otsu, wherein: the noise reduction filtering adopts an adaptive bilateral filter, and the filter weight function is:

[0013] The filtered pixel is calculated as:

[0014] In the formula, W(x,y,i,j) is the joint weight of the center pixel (x,y) and the neighborhood (i,j); (x,y) is the center coordinate of the pixel to be filtered; (i,j) is the coordinate of any pixel in the neighborhood; σ d is the spatial domain standard deviation; σ r is the intensity domain standard deviation; σ n is the noise variance; alpha is the noise sensitivity coefficient; I(i,j) is the original pixel value at the coordinate (i,j); I(x,y) is the original pixel value at the coordinate (x,y); and Omega is the filtering window centered at (x,y).

[0015] As a preferred embodiment, the preprocessed electronic image specifically includes noise reduction filtering, geometric correction and adaptive threshold segmentation based on improved Otsu, wherein: the geometric correction adopts a perspective transformation matrix; specifically:

[0016] wherein:

[0017] wherein: (x, y) is the original center coordinate; (x', y') is the corresponding point coordinate of (x, y) in the transformed image; M is the secondary homography matrix.

[0018] As a preferred embodiment, the electronic image is pre-processed, specifically including noise reduction filtering, geometric correction and adaptive threshold segmentation based on improved Otsu, wherein the adaptive threshold segmentation based on improved Otsu specifically includes the following steps:

[0019] Gradient amplitude calculation:

[0020] wherein:

[0021] wherein: G(x, y) is the gradient amplitude of pixel (x, y); I is the input grayscale image matrix; S x is the Sobel horizontal direction operator; S y is the Sobel vertical direction operator; * is the convolution operator;

[0022] Weighted histogram construction:

[0023] wherein: p(t) is the grayscale weighted histogram; γ is the gradient threshold value; δ() is the Dirac function; t is the candidate segmentation threshold value;

[0024] Improved inter-class variance function by introducing regional contrast weight:

[0025] wherein: is the improved inter-class variance; ω0(t), ω1(t) are the proportions of foreground and background pixels respectively; μ1(t), μ2(t) are the average values of foreground and background grayscale respectively; μ global is the global grayscale average value;

[0026] Optimal threshold calculation:

[0027] wherein: t * is the optimal segmentation threshold value; L is the total number of grayscale levels.

[0028] As a preferred embodiment, the pre-processed image is subjected to data positioning and extraction by using an image recognition algorithm, which includes the fusion application of Canny edge detection, morphological operation and multi-code decoding engine.

[0029] As a preferred embodiment, the data consistency verification adopts a multi-modal fusion verification, and the specific steps are as follows: SIFT feature descriptors of an image are extracted to generate hash values; chip data are read to generate hash values; the hash values of the image feature descriptors are compared with the hash values of the chip data, and if the comparison result is greater than a pre-judgment threshold, manual review is triggered.

[0030] As a preferred embodiment, the tamper-proof storage of the blockchain satisfies:

[0031] Each block contains a Merkle tree root hash of at least 10 operation records;

[0032] The prev_hash field in the block header matches the SHA-256 value of the previous block;

[0033] The asymmetric encryption signature adopts an ECDSA algorithm, and the private key fragments are stored in multiple physical nodes.

[0034] In another aspect, the application also provides a cross-system data collaborative processing system based on image recognition and chip storage, comprising:

[0035] An image acquisition and data extraction module acquires an electronic image of physical medium data through an image acquisition device or a virtual printer, pre-processes the electronic image, and locates and extracts data from the pre-processed image by using an image recognition algorithm;

[0036] A chip interaction module writes the extracted data into a designated sector of an embedded chip, and verifies data consistency by a CRC32 checksum and read-back comparison mechanism;

[0037] A cross-system dynamic adaptation transmission module performs context-aware format adaptation on chip data by a dynamic format conversion model according to the encoding requirements of a target system, and performs automatic data transmission across systems by using an API interface and a SendKeys method;

[0038] A log tracing module generates a full-process operation log, and the operation log contains a timestamp, a chip serial number, a data hash value, and a tamper-proof storage identification of the blockchain, and supports abnormal backtracking and auditing.

[0039] In still another aspect, the application also provides an electronic device having a computer program stored thereon, and the computer program is executed by a processor to implement the cross-system data collaborative processing method based on image recognition and chip storage according to any one of the embodiments of the application.

[0040] In still another aspect, the present application also provides a computer readable medium for storing one or more programs, which when executed by one or more processors, cause the one or more processors to implement the cross-system data collaborative processing method based on image recognition and chip storage as described in any of the embodiments of the present application.

[0041] The present application has the following beneficial effects:

[0042] 1. By fusing adaptive bilateral filtering, perspective transformation correction and improved Otsu segmentation, the image quality of distorted and noisy documents is significantly improved, and the text positioning accuracy is improved by more than 40%;

[0043] 2. Based on the context-aware dynamic format conversion model, the hybrid transmission mode of API interface and SendKeys simulation input is supported, which can adapt to more than 90% of heterogeneous systems;

[0044] 3. The multi-modal verification mechanism of CRC32 verification and SIFT feature hash comparison reduces the missed detection rate to less than 0.01%; combined with blockchain storage and private key fragmentation technology, tamper-proof audit tracking is realized;

[0045] 4. The operation log integrates timestamp, chip serial number, data hash value, and is bound to the blockchain through the Merkle tree, supporting second-level abnormal positioning and improving audit efficiency. BRIEF DESCRIPTION OF DRAWINGS

[0046] 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 of the present application. 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 on the scope. For those skilled in the art, other related drawings can also be obtained without creative labor on the basis of these drawings.

[0047] Fig. 1 is a flowchart of the method of the present application. DETAILED DESCRIPTION

[0048] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, not all. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0049] It should be understood that the step numbers used herein are only for the convenience of description, and are not limited to the execution sequence of the steps.

[0050] It should be understood that the terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the invention. As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.

[0051] The terms “comprising” and “including” indicate the presence of the described feature, whole, step, operation, element and / or component, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or collections thereof.

[0052] The term “and / or” refers to any combination of one or more of the associated listed items, as well as all possible combinations, and includes these combinations.

[0053] Example 1:

[0054] To make the objectives, technical solutions, and advantages of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below in conjunction with specific embodiments of the present application and with reference to Figure 1.

[0055] To address the problems of existing technologies, this invention provides a cross-system data collaborative processing method based on image recognition and chip storage, comprising the following steps:

[0056] Electronic images of physical media data are acquired through image acquisition devices or virtual printers, and the electronic images are preprocessed. Image recognition algorithms are then used to locate and extract data from the preprocessed images. The located and extracted data includes text regions, one-dimensional barcodes, two-dimensional matrix codes, geometric figures, and biometric features.

[0057] The electronic image undergoes preprocessing, specifically including noise reduction filtering, geometric correction, and adaptive threshold segmentation based on an improved Otsu algorithm.

[0058] Among them, the noise reduction filtering adopts an adaptive bilateral filter, and the filter weighting function is:

[0059] The filtered pixels are calculated as follows:

[0060] In the formula: W(x,y,i,j) is the joint weight of the center pixel (x,y) and its neighborhood (i,j); (x,y) is the center coordinate of the pixel to be filtered; (i,j) is the coordinate of any pixel in the neighborhood; σ d σ represents the standard deviation of the spatial domain (default value is 1.5 for a 3×3 neighborhood); r The standard deviation of the intensity domain (taken as 5% of the image dynamic range); σ nis noise variance; a is noise sensitivity coefficient, value range 0.2~0.5; I(i,j) is original pixel value at coordinate (i,j); I(x,y) is original pixel value at coordinate (x,y); Ω is filter window with (x,y) as center.

[0061] Wherein: geometric correction adopts perspective transformation matrix; specifically:

[0062] Wherein:

[0063] Wherein: (x,y) is original center coordinate; (x',y') is corresponding point coordinate of (x,y) in transformed image; M is secondary homography matrix.

[0064] Wherein: adaptive threshold segmentation based on improved Otsu, specific steps are:

[0065] Gradient amplitude calculation:

[0066] Wherein:

[0067] Wherein: G(x,y) is gradient amplitude of pixel (x,y); I is input gray image matrix; S x is Sobel horizontal direction operator; S y is Sobel vertical direction operator; * is convolution operator;

[0068] Weighted histogram construction:

[0069] Wherein: p(t) is gray weighted histogram; γ is gradient threshold, taking 75% quantile of gradient distribution; δ() is Dirac function, used for statistics of gray frequency; t is candidate segmentation threshold;

[0070] Improved inter-class variance function is introduced by introducing regional contrast weight:

[0071] Wherein: is improved inter-class variance; ω0(t), ω1(t) are foreground and background pixel proportion respectively; μ1(t), μ2(t) are foreground and background gray average value respectively; μ global is global gray average value;

[0072] Optimal threshold calculation:

[0073] Wherein: t * is optimal segmentation threshold; L is total number of gray levels.

[0074] The image recognition algorithm is used to locate and extract data from the preprocessed image. The image recognition algorithm includes the fusion application of Canny edge detection, morphological operations and multi-code decoding engine.

[0075] The extracted data is written to a specified sector of the embedded chip, and the data consistency is verified through CRC32 check and readback comparison mechanism.

[0076] The data consistency is verified through CRC32 check and readback comparison mechanism. The data consistency verification adopts multimodal fusion verification. The specific steps are as follows: extract the image SIFT feature descriptor and generate a hash value; read the chip data and generate a hash value; compare the image feature descriptor hash value with the chip data hash value. If it is greater than the pre-approval threshold, in this embodiment, the threshold is set to 5 to trigger manual review.

[0077] Based on the encoding requirements of the target system, the chip data is adapted to a context-aware format through a dynamic format conversion model, and automated data transfer across systems is carried out using API interfaces (such as SwitchToThisWindow) and SendKeys methods.

[0078] Generate a full-process operation log, which includes timestamps, chip serial numbers, data hash values, and blockchain tamper-proof storage identifiers, and supports anomaly backtracking and auditing.

[0079] The blockchain's tamper-proof storage meets the following requirements:

[0080] Each block contains a Merkle root hash of at least 10 operation records;

[0081] The prev_hash field in the block header matches the SHA-256 value of the previous block;

[0082] The asymmetric cryptographic signature uses the ECDSA algorithm, and the private key is stored in fragments across multiple physical nodes.

[0083] Example 2:

[0084] The image acquisition and data extraction module acquires electronic images of physical media data through image acquisition devices or virtual printers, preprocesses the electronic images, and uses image recognition algorithms to locate and extract data from the preprocessed images.

[0085] The chip interaction module writes the extracted data into a designated sector of the embedded chip and verifies the data consistency through CRC32 checksum and readback comparison mechanism.

[0086] The cross-system dynamic adaptation transmission module adapts chip data to a context-aware format based on the encoding requirements of the target system through a dynamic format conversion model, and uses API interfaces and SendKeys methods for automated cross-system data transfer.

[0087] The log traceability module generates a full-process operation log, which includes timestamps, chip serial numbers, data hash values, and blockchain tamper-proof storage identifiers, and supports anomaly backtracking and auditing.

[0088] Example 3:

[0089] This embodiment provides an electronic device on which a computer program is stored. When the computer program is executed by a processor, it implements the cross-system data collaborative processing method based on image recognition and chip storage as described in any embodiment of the present invention.

[0090] Example 4:

[0091] This embodiment provides a computer-readable medium for storing one or more programs, which, when executed by one or more processors, cause the one or more processors to implement the cross-system data collaborative processing method based on image recognition and chip storage as described in any embodiment of the present invention.

[0092] In this application embodiment, "at least one" refers to one or more, and "more than one" refers to two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent the existence of A alone, A and B simultaneously, or B alone. A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one of the following" and similar expressions refer to any combination of these items, including any combination of singular or plural items. For example, at least one of a, b, and c can represent: a, b, c, a and b, a and c, b and c, or a and b and c, where a, b, and c can be single or multiple.

[0093] Those skilled in the art will recognize that the units and algorithm steps described in the embodiments disclosed herein can be implemented using electronic hardware, computer software, or a combination of electronic hardware and software. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0094] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0095] In the several embodiments provided in this application, any function, if implemented as a software functional unit and sold or used as an independent product, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, essentially, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0096] The above description is merely an embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structural or procedural transformations made based on the content of the present invention's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. A cross-system data collaborative processing method based on image recognition and chip storage, characterized in that, Includes the following steps: Electronic images of physical media data are acquired through image acquisition devices or virtual printers, and the electronic images are preprocessed. Image recognition algorithms are then used to locate and extract data from the preprocessed images. The extracted data is written to a specified sector of the embedded chip, and the data consistency is verified through CRC32 check and readback comparison mechanism. Based on the coding requirements of the target system, a dynamic format conversion model is used to perform context-aware format adaptation of chip data, and API interfaces and SendKeys methods are used for automated cross-system data transfer. Generate a full-process operation log, which includes timestamps, chip serial numbers, data hash values, and blockchain tamper-proof storage identifiers, and supports anomaly backtracking and auditing.

2. The cross-system data collaborative processing method based on image recognition and chip storage according to claim 1, characterized in that: The electronic image undergoes preprocessing, specifically including noise reduction filtering, geometric correction, and adaptive threshold segmentation based on an improved Otsu filter. Specifically, the noise reduction filtering employs an adaptive bilateral filter, and the filter weighting function is: The filtered pixels are calculated as follows: In the formula: W(x,y,i,j) is the joint weight of the center pixel (x,y) and its neighborhood (i,j); (x,y) is the center coordinate of the pixel to be filtered; (i,j) is the coordinate of any pixel in the neighborhood; σ d σ represents the standard deviation of the spatial domain. r σ represents the standard deviation of the intensity domain. n Ω represents the noise variance; α represents the noise sensitivity coefficient; I(i,j) represents the original pixel value at coordinate (i,j); I(x,y) represents the original pixel value at coordinate (x,y); and Ω represents the filtering window centered at (x,y).

3. The cross-system data collaborative processing method based on image recognition and chip storage according to claim 1, characterized in that: The electronic image undergoes preprocessing, specifically including noise reduction filtering, geometric correction, and adaptive thresholding based on an improved Otsu's algorithm. Specifically, geometric correction employs a perspective transformation matrix. in: In the formula: (x,y) are the original center coordinates; (x′,y′) are the coordinates of the corresponding point (x,y) in the transformed image; M is the homogeneous transformation matrix.

4. The cross-system data collaborative processing method based on image recognition and chip storage according to claim 1, characterized in that: The electronic image undergoes preprocessing, specifically including noise reduction filtering, geometric correction, and adaptive threshold segmentation based on improved Otsu's algorithm. The adaptive threshold segmentation based on improved Otsu's algorithm comprises the following steps: Gradient magnitude calculation: in: In the formula: G(x,y) is the gradient magnitude of pixel (x,y); I is the input grayscale image matrix; S x S is the Sobel horizontal direction operator; y The vertical operator is the Sobel operator; * represents the convolution operator. Constructing a weighted histogram: In the formula: p(t) is the gray-level weighted histogram; γ is the gradient threshold; δ() is the Dirac function; t is the candidate segmentation threshold; Introducing regional contrast weights to improve the inter-class variance function: In the formula: To improve the inter-class variance; ω0(t) and ω1(t) are the foreground and background pixel ratios, respectively; μ1(t) and μ2(t) are the foreground and background grayscale averages, respectively; μ global This represents the global average grayscale value. Optimal threshold calculation: In the formula: t * is the optimal segmentation threshold; L is the total number of gray levels.

5. The cross-system data collaborative processing system based on image recognition and chip storage according to claim 1, characterized in that: The image recognition algorithm is used to locate and extract data from the preprocessed image. The image recognition algorithm includes the fusion application of Canny edge detection, morphological operations and multi-code decoding engine.

6. The cross-system data collaborative processing method based on image recognition and chip storage according to claim 1, characterized in that: The data consistency is verified through CRC32 check and readback comparison mechanism. The data consistency verification adopts multimodal fusion verification, and the specific steps are as follows: extract the image SIFT feature descriptor and generate a hash value; read the chip data and generate a hash value; compare the image feature descriptor hash value with the chip data hash value. If it is greater than the pre-approval threshold, manual review is triggered.

7. The cross-system data collaborative processing method based on image recognition and chip storage according to claim 1, characterized in that: The blockchain's tamper-proof storage meets the following requirements: Each block contains a Merkle root hash of at least 10 operation records; The prev_hash field in the block header matches the SHA-256 value of the previous block; The asymmetric cryptographic signature uses the ECDSA algorithm, and the private key is stored in fragments across multiple physical nodes.

8. A cross-system data collaborative processing system based on image recognition and chip storage, characterized in that, include: The image acquisition and data extraction module acquires electronic images of physical media data through image acquisition devices or virtual printers, preprocesses the electronic images, and uses image recognition algorithms to locate and extract data from the preprocessed images. The chip interaction module writes the extracted data into a designated sector of the embedded chip and verifies the data consistency through CRC32 checksum and readback comparison mechanism. The cross-system dynamic adaptation transmission module adapts chip data to a context-aware format based on the encoding requirements of the target system through a dynamic format conversion model, and uses API interfaces and SendKeys methods for automated cross-system data transfer. The log traceability module generates a full-process operation log, which includes timestamps, chip serial numbers, data hash values, and blockchain tamper-proof storage identifiers, and supports anomaly backtracking and auditing.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the cross-system data collaborative processing method based on image recognition and chip storage as described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by the processor, the program implements the cross-system data collaborative processing method based on image recognition and chip storage as described in any one of claims 1 to 7.

Images