An intelligent pathological section label identification method, system and medium

Abstract

The application provides a kind of intelligent pathological section label identification method, system and medium, the method includes: receiving the input image of pathological section label, based on the identification mode of pre-set identification mode, the image format corresponding to input image and OCR type configuration string are identified;Call identification mode to execute identification operation to input image, analyze whether the identification result is valid;If the identification result is valid, then based on the identification mode of pre-set identification mode, the identification result is output;If the identification result is invalid, then switch identification mode, secondary identification is carried out to input image, until the identification result is valid, improve the identification efficiency of intelligent pathological section label;Wherein the priority strategy under automatic mode and intelligent back mechanism, combined with image enhancement technology, ensure the normal operation of system under the condition of multiple angles, low definition, information loss etc. Efficient identification of OCR and two-dimensional code is realized through multi-strategy joint identification mechanism.

Description

Technical Field

[0001] This application relates to the field of slide recognition technology, and more specifically, to an intelligent pathological slide label recognition method, system, and medium. Background Technology

[0002] Traditional OCR recognition solutions have extremely high requirements for image quality. The recognition rate drops sharply under conditions such as uneven lighting, low contrast, and blur. They cannot handle rotated or tilted text, require manual preprocessing, have slow recognition speed, poor real-time performance, and low accuracy in recognizing mixed Chinese and English text.

[0003] Traditional QR code recognition solutions rely on open-source libraries such as ZXing, which have limited recognition capabilities, insufficient image preprocessing, inability to handle blurry or noisy images, lack of intelligent backoff mechanisms, and will abandon the recognition if the first recognition result is invalid. They also have poor encoding compatibility and incomplete support for Chinese characters.

[0004] The single recognition mode solution only supports a single recognition method and cannot automatically select the best recognition strategy based on the image content. It lacks a hybrid recognition mechanism and cannot make full use of the advantages of multiple recognition technologies, resulting in a poor user experience and requiring manual selection of the recognition mode. Summary of the Invention

[0005] The purpose of this application is to provide an intelligent pathological slide label recognition method, system, and medium. It achieves efficient recognition of OCR and QR codes through a multi-strategy joint recognition mechanism. The priority strategy and intelligent backoff mechanism in automatic mode, combined with image grading enhancement technology, improve the efficiency of recognition results and make the recognition accuracy more accurate.

[0006] This application also provides an intelligent pathological slide label recognition method, including:

[0007] The system receives an input image of a pathological slide label and identifies the image format and OCR type configuration string corresponding to the input image based on a preset recognition mode. The preset recognition mode includes at least a QR code recognition mode, an OCR recognition mode, and an automatic recognition mode.

[0008] The recognition mode is invoked to perform a recognition operation on the input image, and the validity of the recognition result is analyzed.

[0009] If the recognition result is valid, the recognition result will be output based on the preset recognition mode;

[0010] If the recognition result is invalid, switch the recognition mode and perform a second recognition on the input image until the recognition result is valid.

[0011] Optionally, in the intelligent pathological slide label recognition method described in the embodiments of this application, after receiving the input image of the pathological slide label, it further includes:

[0012] Establish a quality assessment model for pathological slide label images based on big data;

[0013] The input image quality assessment model for pathological slide labels extracts the sharpness, contrast, and noise density of the input image.

[0014] The sharpness, contrast, and noise density of the input image are used to calculate an image quality score based on preset weights.

[0015] Compare the image quality score with a preset score;

[0016] If the image quality score is greater than or equal to the preset score, the original input image will be output directly.

[0017] If the image quality score is less than the preset score, contrast enhancement and Gaussian adaptive threshold binarization are performed to obtain the enhanced input image.

[0018] Optionally, in the intelligent pathological slide label recognition method described in this application embodiment, the recognition method for the QR code recognition mode is as follows:

[0019] The system acquires the input image of the pathological slide label, performs raw image recognition on the input image, and outputs the QR code recognition result directly if the recognition result is valid.

[0020] If the original image recognition result is invalid, the input image is enhanced to obtain an enhanced image;

[0021] Perform preprocessing on the enhanced image and convert the preprocessed enhanced image into a standard format;

[0022] The enhanced image after format conversion is scanned to obtain a scanned image;

[0023] The scanned image is decoded to obtain the decoding result. Garbled characters are detected in the decoding result, and those containing garbled characters are filtered out. Finally, a QR code recognition result that conforms to the encoding format is output.

[0024] Optionally, in the intelligent pathological slide label recognition method described in this application embodiment, the recognition steps of the OCR recognition mode are as follows:

[0025] The OCR recognition model is used to detect text regions in the input image and output the coordinates of the text regions.

[0026] Text direction is detected based on text region coordinates to obtain detection results;

[0027] The text region is rotated and corrected based on the detection results;

[0028] Perform character recognition on the rotated text region and output the initial OCR recognition result.

[0029] Optionally, in the intelligent pathological slide label recognition method described in this application embodiment, the recognition result is output based on a preset recognition pattern, and then the method further includes post-processing of the recognition result, as follows:

[0030] The post-processing includes at least filename validity checking, garbled character detection and filtering;

[0031] Set a confidence threshold to filter out initial OCR recognition results that are below the confidence threshold;

[0032] Configure the string according to the OCR type and check whether the recognition result contains special characters prohibited by the file system;

[0033] If special characters exist, they will be replaced with preset legal characters or deleted to obtain recognition results that meet the file requirements. After filtering out special characters, the final OCR recognition result will be output.

[0034] Optionally, in the intelligent pathological slide label recognition method described in this application embodiment, a priority recognition strategy is executed in automatic mode, as follows:

[0035] The QR code recognition mode is used first to recognize the input image;

[0036] If the QR code recognition result is invalid, it will automatically switch to OCR recognition mode to perform recognition.

[0037] Secondly, embodiments of this application provide an intelligent pathological slide label recognition system. The system includes a memory and a processor. The memory includes a program for an intelligent pathological slide label recognition method. When the program for the intelligent pathological slide label recognition method is executed by the processor, it performs the following steps:

[0038] The system receives an input image of a pathological slide label and identifies the image format and OCR type configuration string corresponding to the input image based on a preset recognition mode. The preset recognition mode includes at least a QR code recognition mode, an OCR recognition mode, and an automatic recognition mode.

[0039] The system invokes the recognition mode to perform recognition operations on the input image and analyzes the recognition results.

[0040] The validity of the identification results is analyzed based on their effectiveness.

[0041] If the recognition result is valid, the recognition result will be output based on the preset recognition mode;

[0042] If the recognition result is invalid, the recognition mode is switched and the input image is recognized a second time until the recognition result is valid, thereby improving the recognition efficiency of the intelligent pathological slide label.

[0043] Optionally, in the intelligent pathological slide label recognition system described in this application embodiment, after receiving the input image of the pathological slide label, it further includes:

[0044] Establish a quality assessment model for pathological slide label images based on big data;

[0045] The input image quality assessment model for pathological slide labels extracts the sharpness, contrast, and noise density of the input image.

[0046] The sharpness, contrast, and noise density of the input image are used to calculate an image quality score based on preset weights.

[0047] Compare the image quality score with a preset score;

[0048] If the image quality score is greater than or equal to the preset score, the original input image will be output directly.

[0049] If the image quality score is less than the preset score, contrast enhancement and Gaussian adaptive threshold binarization are performed to obtain the enhanced input image.

[0050] Optionally, in the intelligent pathological slide label recognition system described in this application embodiment, the recognition method for the QR code recognition mode is as follows:

[0051] The system acquires the input image of the pathological slide label, performs raw image recognition on the input image, and outputs the QR code recognition result directly if the recognition result is valid.

[0052] If the original image recognition result is invalid, the input image is enhanced to obtain an enhanced image;

[0053] Perform preprocessing on the enhanced image and convert the preprocessed enhanced image into a standard format;

[0054] The enhanced image after format conversion is scanned to obtain a scanned image;

[0055] The scanned image is decoded to obtain the decoding result. Garbled characters are detected in the decoding result, and those containing garbled characters are filtered out. Finally, a QR code recognition result that conforms to the encoding format is output.

[0056] Thirdly, embodiments of this application also provide a computer-readable storage medium, which includes a program for intelligent pathological slide label recognition, wherein when the program is executed by a processor, it implements the steps of the intelligent pathological slide label recognition method as described in any of the preceding claims.

[0057] As can be seen from the above, the intelligent pathological slide label recognition method, system, and medium provided in this application receive an input image of a pathological slide label and recognize the image format and OCR type configuration string corresponding to the input image based on a preset recognition mode. The preset recognition mode includes at least a QR code recognition mode, an OCR recognition mode, and an automatic recognition mode. The recognition mode is invoked to perform a recognition operation on the input image, and the validity of the recognition result is analyzed. If the recognition result is valid, the recognition result is output based on the preset recognition mode. If the recognition result is invalid, the recognition mode is switched, and the input image is recognized a second time until the recognition result is valid. The multi-strategy joint recognition mechanism achieves efficient recognition of OCR and QR codes. The priority strategy and intelligent backoff mechanism in the automatic mode, combined with image grading enhancement technology, improve the efficiency of recognition results and the accuracy of recognition. Attached Figure Description

[0058] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments of this application will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0059] Figure 1 A flowchart of the intelligent pathological slide label recognition method provided in the embodiments of this application;

[0060] Figure 2 A flowchart illustrating the input image quality assessment of the intelligent pathological slide label recognition method provided in this application embodiment;

[0061] Figure 3 This is a schematic diagram of the QR code recognition process of the intelligent pathological slide label recognition method provided in this application embodiment. Detailed Implementation

[0062] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of the embodiments. The components of the embodiments of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely represents selected embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0063] It should be noted that similar reference numerals 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. Furthermore, in the description of this application, the terms "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0064] Please refer to Figure 1 , Figure 1 This is a flowchart of an intelligent pathological slide label recognition method according to some embodiments of this application. The intelligent pathological slide label recognition method is used in a terminal device and includes the following steps:

[0065] S101, receive the input image of the pathological slide label, and identify the image format and OCR type configuration string corresponding to the input image based on the preset recognition mode. The preset recognition mode includes at least QR code recognition mode, OCR recognition mode and automatic recognition mode.

[0066] S102, invoke the recognition mode to perform a recognition operation on the input image, and analyze whether the recognition result is valid;

[0067] S103, If the recognition result is valid, output the recognition result based on the preset recognition mode;

[0068] S104 If the recognition result is invalid, switch the recognition mode and perform a second recognition on the input image until the recognition result is valid, thereby improving the recognition efficiency of the intelligent pathological slide label.

[0069] It should be noted that this application achieves intelligent fusion of OCR recognition and QR code recognition. It can automatically select the best recognition strategy based on image features, provide an automatic mode, prioritize QR code recognition, and automatically switch to OCR recognition after failure, maximizing the efficiency of recognition results. It supports multiple recognition modes (OCR, QR code, automatic) to meet the needs of different application scenarios.

[0070] Please refer to Figure 2 , Figure 2 This is a flowchart illustrating the input image quality assessment of an intelligent pathological slide label recognition method according to some embodiments of this application. According to an embodiment of the present invention, after receiving the input image of the pathological slide label, the method further includes:

[0071] S201, Establishing a quality assessment model for pathological slide label images based on big data;

[0072] S202, the input image quality assessment model of pathological slide labels, extracts the sharpness, contrast and noise density of the input image;

[0073] S203, The sharpness, contrast and noise density of the input image are used to calculate the image quality score according to the preset weights;

[0074] S204, compare the image quality score with the preset score;

[0075] S205: If the image quality score is greater than or equal to the preset score, the original input image is directly output; if the image quality score is less than the preset score, contrast enhancement and Gaussian adaptive threshold binarization are performed to obtain the enhanced input image.

[0076] It should be noted that a special image enhancement algorithm was designed for unclear QR codes, which significantly improves the recognition efficiency. Contrast enhancement and adaptive threshold binarization are used to optimize image quality.

[0077] Intelligent rollback mechanism: When the original image recognition result is invalid, image enhancement is automatically performed before attempting recognition again.

[0078] Please refer to Figure 3 , Figure 3 This is a schematic diagram illustrating the QR code recognition process of an intelligent pathological slide label recognition method according to some embodiments of this application. According to embodiments of the present invention, the QR code recognition method is as follows:

[0079] S301, acquire the input image of the pathological slide label, perform raw image recognition on the input image, and if the recognition result is valid, directly output the QR code recognition result;

[0080] S302, If the original image recognition result is invalid, perform enhancement processing on the input image to obtain an enhanced image;

[0081] S303 performs preprocessing on the enhanced image and converts the preprocessed enhanced image into a standard format;

[0082] S304, The enhanced image after format conversion is scanned to obtain a scanned image;

[0083] S305 decodes the scanned image to obtain the decoding result, performs garbled character detection on the decoding result, filters out decoding results containing garbled characters, and outputs a QR code recognition result that conforms to the encoding format.

[0084] According to an embodiment of the present invention, the recognition steps of the OCR recognition pattern are as follows:

[0085] The OCR recognition model is used to detect text regions in the input image and output the coordinates of the text regions.

[0086] Text direction is detected based on text region coordinates to obtain detection results;

[0087] The text region is rotated and corrected based on the detection results;

[0088] Perform character recognition on the rotated text region and output the initial OCR recognition result.

[0089] It should be noted that the OCR type configuration string is a 4-character string, which controls the recognition of numbers, English letters, Chinese characters, and spaces respectively.

[0090] The OCR recognition model is a deep learning model based on PaddleOCR, which supports mixed Chinese and English recognition. It integrates three-stage processing of text detection, classification and recognition, has high recognition accuracy, supports automatic text direction detection and rotation correction, and has configurable recognition parameters. It supports flexible combination recognition of numbers, English, Chinese and spaces.

[0091] According to an embodiment of the present invention, a recognition result is output based on a preset recognition pattern, and the method further includes post-processing of the recognition result, the steps of which are as follows:

[0092] Post-processing should include at least filename validity checks, garbled character detection, and filtering;

[0093] Set a confidence threshold to filter out initial OCR recognition results that are below the confidence threshold;

[0094] Configure the string according to the OCR type and check whether the recognition result contains special characters prohibited by the file system;

[0095] If special characters exist, they will be replaced with preset legal characters or deleted to obtain recognition results that meet the file requirements. After filtering out special characters, the final OCR recognition result will be output.

[0096] According to an embodiment of the present invention, a priority identification strategy is executed in automatic mode, specifically as follows:

[0097] The QR code recognition mode is used first to recognize the input image;

[0098] If the QR code recognition result is invalid, it will automatically switch to OCR recognition mode to perform recognition.

[0099] According to an embodiment of the present invention, the method further includes: if the image quality score is less than a preset threshold, then a graded enhancement strategy is executed, specifically as follows:

[0100] Noise levels are analyzed based on noise density. Noise levels include:

[0101] Low noise level (noise density < 0.2): Only contrast enhancement and Gaussian adaptive threshold binarization are performed;

[0102] Medium noise level (0.2≤noise density≤0.5): First perform median filtering for noise reduction, then perform contrast enhancement and Gaussian adaptive threshold binarization.

[0103] High noise level (noise density > 0.5): Perform bilateral filtering denoising, contrast enhancement, Gaussian adaptive threshold binarization, and morphological dilation processing in sequence;

[0104] The enhanced image is output to the recognition module, and the enhancement strategy parameters are recorded for subsequent recognition result tracking.

[0105] It should be noted that by analyzing different noise densities and applying different filtering and denoising methods, the denoising effect can be improved, making the input image clearer and the recognition result more accurate.

[0106] Secondly, embodiments of this application provide an intelligent pathological slide label recognition system. The system includes a memory and a processor. The memory contains a program for an intelligent pathological slide label recognition method. When the processor executes the program for the intelligent pathological slide label recognition method, it performs the following steps:

[0107] The system receives input images of pathological slide labels and identifies the corresponding image format and OCR type configuration string based on preset recognition modes. The preset recognition modes include at least QR code recognition mode, OCR recognition mode, and automatic recognition mode.

[0108] The recognition mode is invoked to perform a recognition operation on the input image, and the validity of the recognition result is analyzed.

[0109] If the recognition result is valid, the recognition result will be output based on the preset recognition mode;

[0110] If the recognition result is invalid, switch the recognition mode and perform a second recognition on the input image until the recognition result is valid.

[0111] It should be noted that this application achieves intelligent fusion of OCR recognition and QR code recognition. It can automatically select the best recognition strategy based on image features, provide an automatic mode, prioritize QR code recognition, and automatically switch to OCR recognition after failure, maximizing the efficiency of recognition results. It supports multiple recognition modes (OCR, QR code, automatic) to meet the needs of different application scenarios.

[0112] According to an embodiment of the present invention, after receiving the input image of the pathological slide label, the method further includes:

[0113] Establish a quality assessment model for pathological slide label images based on big data;

[0114] The input image quality assessment model for pathological slide labels extracts the sharpness, contrast, and noise density of the input image.

[0115] The sharpness, contrast, and noise density of the input image are used to calculate an image quality score based on preset weights.

[0116] Compare the image quality score with a preset score;

[0117] If the image quality score is greater than or equal to the preset score, the original input image will be output directly.

[0118] If the image quality score is less than the preset score, contrast enhancement and Gaussian adaptive threshold binarization are performed to obtain the enhanced input image.

[0119] It should be noted that a special image enhancement algorithm was designed for unclear QR codes, which significantly improves the recognition efficiency. Contrast enhancement and adaptive threshold binarization are used to optimize image quality.

[0120] Intelligent rollback mechanism: When the original image recognition result is invalid, image enhancement is automatically performed before attempting recognition again.

[0121] According to an embodiment of the present invention, the method for recognizing a QR code recognition pattern is as follows:

[0122] The system acquires the input image of the pathological slide label, performs raw image recognition on the input image, and outputs the QR code recognition result directly if the recognition result is valid.

[0123] If the original image recognition result is invalid, the input image is enhanced to obtain an enhanced image;

[0124] Perform preprocessing on the enhanced image and convert the preprocessed enhanced image into a standard format;

[0125] The enhanced image after format conversion is scanned to obtain a scanned image;

[0126] The scanned image is decoded to obtain the decoding result. Garbled characters are detected in the decoding result, and those containing garbled characters are filtered out. Finally, a QR code recognition result that conforms to the encoding format is output.

[0127] According to an embodiment of the present invention, the recognition steps of the OCR recognition pattern are as follows:

[0128] The OCR recognition model is used to detect text regions in the input image and output the coordinates of the text regions.

[0129] Text direction is detected based on text region coordinates to obtain detection results;

[0130] The text region is rotated and corrected based on the detection results;

[0131] Perform character recognition on the rotated text region and output the initial OCR recognition result.

[0132] It should be noted that the OCR type configuration string is a 4-character string, which controls the recognition of numbers, English letters, Chinese characters, and spaces respectively.

[0133] The OCR recognition model is a deep learning model based on PaddleOCR, which supports mixed Chinese and English recognition. It integrates three-stage processing of text detection, classification and recognition, has high recognition accuracy, supports automatic text direction detection and rotation correction, and has configurable recognition parameters. It supports flexible combination recognition of numbers, English, Chinese and spaces.

[0134] According to an embodiment of the present invention, a recognition result is output based on a preset recognition pattern, and the method further includes post-processing of the recognition result, the steps of which are as follows:

[0135] Post-processing should include at least filename validity checks, garbled character detection, and filtering;

[0136] Set a confidence threshold to filter out initial OCR recognition results that are below the confidence threshold;

[0137] Configure the string according to the OCR type and check whether the recognition result contains special characters prohibited by the file system;

[0138] If special characters exist, they will be replaced with preset legal characters or deleted to obtain recognition results that meet the file requirements. After filtering out special characters, the final OCR recognition result will be output.

[0139] According to an embodiment of the present invention, a priority identification strategy is executed in automatic mode, specifically as follows:

[0140] The QR code recognition mode is used first to recognize the input image;

[0141] If the QR code recognition result is invalid, it will automatically switch to OCR recognition mode to perform recognition.

[0142] According to an embodiment of the present invention, the method further includes: if the image quality score is less than a preset threshold, then a graded enhancement strategy is executed, specifically as follows:

[0143] Noise levels are analyzed based on noise density. Noise levels include:

[0144] Low noise level (noise density < 0.2): Only contrast enhancement and Gaussian adaptive threshold binarization are performed;

[0145] Medium noise level (0.2≤noise density≤0.5): First perform median filtering for noise reduction, then perform contrast enhancement and Gaussian adaptive threshold binarization.

[0146] High noise level (noise density > 0.5): Perform bilateral filtering denoising, contrast enhancement, Gaussian adaptive threshold binarization, and morphological dilation processing in sequence;

[0147] The enhanced image is output to the recognition module, and the enhancement strategy parameters are recorded for subsequent recognition result tracking.

[0148] It should be noted that by analyzing different noise densities and applying different filtering and denoising methods, the denoising effect can be improved, making the input image clearer and the recognition result more accurate.

[0149] A third aspect of the present invention provides a computer-readable storage medium including a program for an intelligent pathological slide label recognition method, wherein when the program is executed by a processor, it implements the steps of the intelligent pathological slide label recognition method as described in any of the above claims.

[0150] This invention discloses an intelligent pathological slide label recognition method, system, and medium. It receives an input image of a pathological slide label and identifies the corresponding image format and OCR type configuration string based on a preset recognition mode. The preset recognition mode includes at least a QR code recognition mode, an OCR recognition mode, and an automatic recognition mode. The system calls the recognition mode to perform a recognition operation on the input image and analyzes the validity of the recognition result. If the recognition result is valid, it outputs the recognition result based on the preset recognition mode. If the recognition result is invalid, it switches the recognition mode and performs a second recognition on the input image until the recognition result is valid. The system achieves efficient recognition of OCR and QR codes through a multi-strategy joint recognition mechanism. The priority strategy and intelligent backoff mechanism in automatic mode, combined with image grading enhancement technology, improve the efficiency and accuracy of the recognition results.

[0151] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods, such as: multiple units or components can be combined, or integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the various components shown or discussed can be through some interfaces, and the indirect coupling or communication connection between devices or units can be electrical, mechanical, or other forms.

[0152] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units. They may be located in one place or distributed across multiple network units. Some or all of the units may be selected to achieve the purpose of this embodiment according to actual needs.

[0153] In addition, in the various embodiments of the present invention, each functional unit can be integrated into one processing unit, or each unit can be a separate unit, or two or more units can be integrated into one unit; the integrated unit can be implemented in hardware or in the form of hardware plus software functional units.

[0154] Those skilled in the art will understand that all or part of the steps of the above method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a readable storage medium. When the program is executed, it performs the steps of the above method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as mobile storage devices, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0155] Alternatively, if the integrated units of the present invention are implemented as software functional modules and sold or used as independent products, they can also be stored in a readable storage medium. Based on this understanding, the technical solutions of the embodiments of the present invention, or the parts that contribute to the prior art, can be embodied in the form of a software product. This software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as mobile storage devices, ROM, RAM, magnetic disks, or optical disks.

Claims

1. A method for intelligent identification of pathological slide labels, characterized in that, include: The system receives an input image of a pathological slide label and identifies the image format and OCR type configuration string corresponding to the input image based on a preset recognition mode. The preset recognition mode includes at least a QR code recognition mode, an OCR recognition mode, and an automatic recognition mode. The recognition mode is invoked to perform a recognition operation on the input image, and the validity of the recognition result is analyzed. If the recognition result is valid, the recognition result will be output based on the preset recognition mode; If the recognition result is invalid, switch the recognition mode and perform a second recognition on the input image until the recognition result is valid.

2. The intelligent pathological slide label recognition method according to claim 1, characterized in that, After receiving the input image of the pathology slide label, the process also includes: Establish a quality assessment model for pathological slide label images based on big data; The input image quality assessment model for pathological slide labels extracts the sharpness, contrast, and noise density of the input image. The sharpness, contrast, and noise density of the input image are used to calculate an image quality score based on preset weights. Compare the image quality score with a preset score; If the image quality score is greater than or equal to the preset score, the original input image will be output directly. If the image quality score is less than the preset score, contrast enhancement and Gaussian adaptive threshold binarization are performed to obtain the enhanced input image.

3. The intelligent pathological slide label recognition method according to claim 2, characterized in that, The recognition method for the QR code recognition mode is as follows: The system acquires the input image of the pathological slide label, performs raw image recognition on the input image, and outputs the QR code recognition result directly if the recognition result is valid. If the original image recognition result is invalid, the input image is enhanced to obtain an enhanced image; Perform preprocessing on the enhanced image and convert the preprocessed enhanced image into a standard format; The enhanced image after format conversion is scanned to obtain a scanned image; The scanned image is decoded to obtain the decoding result. Garbled characters are detected in the decoding result, and those containing garbled characters are filtered out. Finally, a QR code recognition result that conforms to the encoding format is output.

4. The intelligent pathological slide label recognition method according to claim 3, characterized in that, The recognition steps for the OCR recognition mode are as follows: The OCR recognition model is used to detect text regions in the input image and output the coordinates of the text regions. Text direction is detected based on text region coordinates to obtain detection results; The text region is rotated and corrected based on the detection results; Perform character recognition on the rotated text region and output the initial OCR recognition result.

5. The intelligent pathological slide label recognition method according to claim 4, characterized in that, The recognition result is output based on the preset recognition pattern, and the post-processing of the recognition result is also included, with the following steps: The post-processing includes at least filename validity checking, garbled character detection and filtering; Set a confidence threshold to filter out initial OCR recognition results that are below the confidence threshold; Configure the string according to the OCR type and check whether the recognition result contains special characters prohibited by the file system; If special characters exist, they will be replaced with preset legal characters or deleted to obtain recognition results that meet the file requirements. After filtering out special characters, the final OCR recognition result will be output.

6. The intelligent pathological slide label recognition method according to claim 5, characterized in that, In automatic mode, a priority identification strategy is executed as follows: The QR code recognition mode is used first to recognize the input image; If the QR code recognition result is invalid, it will automatically switch to OCR recognition mode to perform recognition.

7. An intelligent pathological slide label recognition system, characterized in that, The system includes a memory and a processor. The memory contains a program for an intelligent pathological slide label recognition method. When the program for the intelligent pathological slide label recognition method is executed by the processor, it performs the following steps: The system receives an input image of a pathological slide label and identifies the image format and OCR type configuration string corresponding to the input image based on a preset recognition mode. The preset recognition mode includes at least a QR code recognition mode, an OCR recognition mode, and an automatic recognition mode. The recognition mode is invoked to perform a recognition operation on the input image, and the validity of the recognition result is analyzed. If the recognition result is valid, the recognition result will be output based on the preset recognition mode; If the recognition result is invalid, switch the recognition mode and perform a second recognition on the input image until the recognition result is valid.

8. The intelligent pathological slide label recognition system according to claim 7, characterized in that, After receiving the input image of the pathology slide label, the process also includes: Establish a quality assessment model for pathological slide label images based on big data; The input image quality assessment model for pathological slide labels extracts the sharpness, contrast, and noise density of the input image. The sharpness, contrast, and noise density of the input image are used to calculate an image quality score based on preset weights. Compare the image quality score with a preset score; If the image quality score is greater than or equal to the preset score, the original input image will be output directly. If the image quality score is less than the preset score, contrast enhancement and Gaussian adaptive threshold binarization are performed to obtain the enhanced input image.

9. The intelligent pathological slide label recognition system according to claim 8, characterized in that, The recognition method for the QR code recognition mode is as follows: The system acquires the input image of the pathological slide label, performs raw image recognition on the input image, and outputs the QR code recognition result directly if the recognition result is valid. If the original image recognition result is invalid, the input image is enhanced to obtain an enhanced image; Perform preprocessing on the enhanced image and convert the preprocessed enhanced image into a standard format; The enhanced image after format conversion is scanned to obtain a scanned image; The scanned image is decoded to obtain the decoding result. Garbled characters are detected in the decoding result, and those containing garbled characters are filtered out. Finally, a QR code recognition result that conforms to the encoding format is output.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a program for intelligent pathological slide label recognition, which, when executed by a processor, implements the steps of the intelligent pathological slide label recognition method as described in any one of claims 1 to 6.

Images