Medicine management classification method of medical system and related device

Abstract

The embodiment of the invention relates to the technical field of data processing, and provides a medicine classification method of a medical system and a related device, and the method comprises the steps: obtaining package bar code image data and specification image data of a newly stored medicine; performing image recognition on the package bar code image data to obtain first medicine classification information; performing character classification information identification on the specification image data to obtain second medicine classification information; comparing the first medicine classification information with the second medicine classification information to obtain a medicine classification information comparison result; and determining a drug classification result according to a drug classification information comparison result, so that auxiliary judgment can be performed in combination with identification of a drug specification and correct judgment of a package bar code in a drug package bar code data identification process, and the technical problem of drug classification error caused by bar code misidentification due to overexposure is avoided.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a drug management classification method and related devices for a medical system. Background Art

[0002] Image recognition technology has been gradually applied to pharmaceutical packaging management in the medical field, enabling automated classification by capturing images of drug packaging. Mainstream solutions utilize computer vision technology to collect and analyze the images and text on packaging, combined with methods such as barcode scanning to identify drug attributes. This helps medical institutions quickly complete drug warehousing, sorting, and inventory management, significantly improving operational efficiency and reducing manual errors, becoming a crucial aid in modern medical supply management.

[0003] However, in practice, complex lighting environments often lead to misjudgments by the recognition system. For example, in strong direct sunlight or on metal packaging, the barcode area on the drug packaging becomes a white spot due to overexposure, and the specular reflection from the packaging surface blurs and distorts the pattern and text, making it difficult to accurately extract effective information. This leads to insufficient drug classification accuracy and a higher error rate. Summary of the Invention

[0004] The embodiment of the present application provides a drug management classification method and related devices for a medical system, which can obtain first drug classification information and second drug classification information based on packaging barcode image data and instruction manual image data, and then determine the drug classification result based on the comparison result of the first drug classification information and the second drug classification information, thereby improving the accuracy of drug classification recognition and reducing the error rate of drug classification recognition.

[0005] A first aspect of an embodiment of the present application provides a method for classifying medicines in a medical system, the method comprising:

[0006] Obtain the packaging barcode image data and instruction sheet image data of newly-entered drugs;

[0007] Performing image recognition on the package barcode image data to obtain first drug classification information;

[0008] Performing text classification information recognition on the instruction manual image data to obtain second drug classification information;

[0009] Comparing the first drug classification information with the second drug classification information to obtain a drug classification information comparison result;

[0010] The drug classification result is determined based on the comparison result of the drug classification information.

[0011] In a possible implementation, performing image recognition on the package barcode image data to obtain the first drug classification information includes:

[0012] Performing multi-angle feature fusion processing on the packaging barcode image data to obtain three-dimensional reconstructed barcode area image data;

[0013] Performing image enhancement processing on the barcode area image data to obtain enhanced barcode area image data;

[0014] Inputting the enhanced barcode area image data into a barcode corner point coordinate recognition model to obtain a barcode corner point coordinate recognition result;

[0015] Determining the target barcode area according to the barcode corner point coordinate recognition result;

[0016] The target barcode area is decoded to obtain first drug classification information.

[0017] In a possible implementation, performing text classification information recognition on the instruction sheet image data to obtain the second drug classification information includes:

[0018] extracting text information and coordinate information from the instruction manual image data;

[0019] aligning the text information according to the coordinate information to obtain aligned text information;

[0020] Performing drug classification key field recognition on the aligned text information to obtain drug classification key field recognition results;

[0021] The second drug classification information is determined based on the identification result of the drug classification key field.

[0022] In a possible implementation, determining the second drug classification information according to the drug classification key field identification result includes:

[0023] The regional attention mechanism is used to extract the key fields of the drug classification recognition results to obtain the approval number, indication and pharmacological classification description;

[0024] According to the approval number, indication and pharmacological classification description, a drug-classification knowledge graph is used for matching to obtain the second drug classification information.

[0025] In a possible implementation, determining a drug classification result based on the comparison result of the drug classification information includes:

[0026] If the information indicated by the comparison result is that the drug classification information is consistent, determining that the drug classification result is the drug classification information determined jointly by the first drug classification information and the second drug classification information as the drug classification result;

[0027] If the comparison result indicates that the drug classification information is inconsistent, an adaptive threshold segmentation algorithm is used to separate the package barcode image data to obtain separated barcode data;

[0028] Using a multi-frame dynamic compensation algorithm to repair the separated barcode data to obtain repaired barcode data;

[0029] Decoding the repaired barcode data to obtain third drug classification information;

[0030] Comparing the third drug classification information with the second drug classification information to obtain a secondary comparison result;

[0031] If the secondary comparison result indicates that the drug classification information is consistent, the drug classification information jointly determined by the second drug classification information and the third drug classification information is confirmed as the drug classification result.

[0032] In this example, by obtaining the packaging barcode image data and instruction manual image data of the newly-entered drugs, image recognition is first performed on the packaging barcode image data to obtain the first drug classification information, and then text classification information recognition is performed on the instruction manual image data to obtain the second drug classification information. The first drug classification information is then compared with the second drug classification information. Based on the comparison result, the accurate drug classification result is determined. In the process of recognizing the drug packaging barcode data, the recognition of the drug instruction manual can be combined to assist in the correct judgment of the packaging barcode, thereby avoiding the technical problem of misrecognition of the barcode due to overexposure, which leads to incorrect drug classification.

[0033] A second aspect of an embodiment of the present application provides a drug classification device for a medical system, the device comprising:

[0034] An acquisition unit, used to acquire the package barcode image data and instruction sheet image data of newly-entered drugs;

[0035] a first processing unit, configured to perform image recognition on the package barcode image data to obtain first drug classification information;

[0036] a second processing unit, configured to perform text classification information recognition on the instruction manual image data to obtain second drug classification information;

[0037] a third processing unit, configured to compare the first drug classification information with the second drug classification information to obtain a drug classification information comparison result;

[0038] The determination unit is used to determine the drug classification result according to the comparison result of the drug classification information.

[0039] In a possible implementation, in the aspect of performing image recognition on the package barcode image data to obtain the first drug classification information, the first processing unit is configured to:

[0040] Performing multi-angle feature fusion processing on the packaging barcode image data to obtain three-dimensional reconstructed barcode area image data;

[0041] Performing image enhancement processing on the barcode area image data to obtain enhanced barcode area image data;

[0042] Inputting the enhanced barcode area image data into a barcode corner point coordinate recognition model to obtain a barcode corner point coordinate recognition result;

[0043] Determining the target barcode area according to the barcode corner point coordinate recognition result;

[0044] The target barcode area is decoded to obtain first drug classification information.

[0045] In a possible implementation, in the aspect of performing text classification information recognition on the instruction sheet image data to obtain the second drug classification information, the second processing unit is configured to:

[0046] extracting text information and coordinate information from the instruction manual image data;

[0047] aligning the text information according to the coordinate information to obtain aligned text information;

[0048] Performing drug classification key field recognition on the aligned text information to obtain drug classification key field recognition results;

[0049] The second drug classification information is determined based on the identification result of the drug classification key field.

[0050] In a possible implementation, in determining the second drug classification information according to the drug classification key field identification result, the second processing unit is configured to:

[0051] The regional attention mechanism is used to extract the key fields of the drug classification recognition results to obtain the approval number, indication and pharmacological classification description;

[0052] According to the approval number, indication and pharmacological classification description, a drug-classification knowledge graph is used for matching to obtain the second drug classification information.

[0053] In a possible implementation, in determining the drug classification result based on the drug classification information comparison result, the determining unit is configured to:

[0054] If the information indicated by the comparison result is that the drug classification information is consistent, determining that the drug classification result is the drug classification information determined jointly by the first drug classification information and the second drug classification information as the drug classification result;

[0055] If the comparison result indicates that the drug classification information is inconsistent, an adaptive threshold segmentation algorithm is used to separate the package barcode image data to obtain separated barcode data;

[0056] Using a multi-frame dynamic compensation algorithm to repair the separated barcode data to obtain repaired barcode data;

[0057] Decoding the repaired barcode data to obtain third drug classification information;

[0058] Comparing the third drug classification information with the second drug classification information to obtain a secondary comparison result;

[0059] If the secondary comparison result indicates that the drug classification information is consistent, the drug classification information jointly determined by the second drug classification information and the third drug classification information is confirmed as the drug classification result.

[0060] A third aspect of an embodiment of the present application provides a terminal, comprising a processor, an input device, an output device, and a memory, wherein the processor, input device, output device, and memory are interconnected, wherein the memory is used to store a computer program, the computer program comprises program instructions, and the processor is configured to call the program instructions and execute the step instructions of the drug classification method of the medical system in the first aspect of the embodiment of the present application.

[0061] The fourth aspect of an embodiment of the present application provides a computer-readable storage medium, wherein the above-mentioned computer-readable storage medium stores a computer program for electronic data exchange, wherein the above-mentioned computer program enables a computer to execute part or all of the steps described in the drug classification method of the medical system in the first aspect of the embodiment of the present application.

[0062] A fifth aspect of the embodiments of the present application provides a computer program product, wherein the computer program product includes a non-transitory computer-readable storage medium storing a computer program, wherein the computer program is operable to cause a computer to execute some or all of the steps described in the drug classification method for a medical system in the first aspect of the embodiments of the present application. The computer program product may be a software installation package. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0064] Figure 1 A schematic diagram of the overall process of a drug classification method for a medical system is provided for an embodiment of the present application;

[0065] Figure 2 A schematic structural diagram of a medicine classification device for a medical system is provided for an embodiment of the present application;

[0066] Figure 3 A schematic diagram of the structure of a terminal provided in an embodiment of the present application;

[0067] Reference numerals:

[0068] 1-acquisition unit; 2-first processing unit; 3-second processing unit; 4-third processing unit; 5-determination unit. DETAILED DESCRIPTION

[0069] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0070] The terms "first," "second," and the like in the specification and claims of this application and the accompanying drawings are used to distinguish between different objects, not to describe a particular order. Furthermore, the terms "including," "having," and any variations thereof, are intended to cover non-exclusive inclusions. For example, a process, method, system, product, or apparatus comprising a series of steps or elements is not limited to the listed steps or elements but may optionally include steps or elements not listed, or may optionally include other steps or elements inherent to the process, method, product, or apparatus.

[0071] References to "embodiments" in this application mean that a particular feature, structure, or characteristic described in connection with the embodiment may be included in at least one embodiment of the application. The appearance of this phrase in various places in the specification does not necessarily refer to the same embodiment, nor does it constitute an independent or alternative embodiment that is mutually exclusive of other embodiments. It is understood, both explicitly and implicitly, by those skilled in the art that the embodiments described in this application may be combined with other embodiments.

[0072] In order to better understand the drug classification method of a medical system provided in an embodiment of the present application, the following first briefly introduces the scenario of applying the drug classification method of the medical system. In the process of the drug sorting system, new drugs entering the warehouse need to be classified. However, in actual operation, the complex lighting environment often causes the recognition system to misjudge. For example, in the scenario of direct strong light or metal packaging, the barcode area of ​​the drug packaging becomes a white spot due to overexposure, and the mirror reflection of the packaging surface blurs and deforms the pattern and text, making it difficult to accurately extract effective information. The accuracy of drug classification and recognition is insufficient, resulting in an increased error rate in drug classification.

[0073] The drug management classification method of the medical system is applied to the drug management classification device of the medical system. Figure 1 The figure shows a schematic diagram of the overall process of a drug management classification method for a medical system. Figure 1 Shown, including:

[0074] S1. Obtain the packaging barcode image data and instruction sheet image data of the newly-entered medicine.

[0075] When new drugs are put into storage, the image acquisition device of the drug sorting system can be used to capture images of the outer packaging of the drugs and the instructions in the drug packaging box to obtain packaging barcode image data and instruction manual image data.

[0076] S2. Perform image recognition on the package barcode image data to obtain first drug classification information.

[0077] Among them, when obtaining the packaging barcode image data, the outer packaging of the medicine can be photographed at multiple angles to obtain multi-angle packaging barcode image data, and then the barcode area image data can be reconstructed based on the multi-angle packaging barcode image. After obtaining the reconstructed barcode area image data, the reconstructed barcode area image data can be input into the barcode corner coordinate recognition model to obtain the barcode corner coordinate recognition result. Finally, based on the barcode corner coordinate recognition result, the target barcode area is obtained, and decoding processing can be performed based on the target barcode area to finally obtain the first drug classification information.

[0078] S3. Perform text classification information recognition on the instruction manual image data to obtain second drug classification information.

[0079] Among them, when identifying text classification information on the instruction manual image data, text extraction technology can be used in combination with establishing coordinates to obtain text information and coordinate information, and then the text information can be aligned according to the coordinate information to obtain aligned text information. After obtaining the aligned text information, OCR technology can be used to identify the drug classification key fields of the aligned text information to obtain the drug classification key field recognition results. Finally, based on the drug classification key field recognition results, the second drug classification information is determined.

[0080] S4. Compare the first drug classification information with the second drug classification information to obtain a drug classification information comparison result.

[0081] Among them, after obtaining the image data of the drug outer packaging and the image data of the drug instructions, the comparison result of the drug classification information can include two results. The first is the successful identification of the drug information. Therefore, under the first result, the drug can be classified according to the first result. The second is the failure to successfully identify the drug information. The second result includes two results. 2.1 The result is that no drug classification information is identified from the barcode because of the overexposure problem. 2.2 The result is that the drug information is identified from the barcode. Overexposure causes the barcode to be recognized by the system as another barcode, but the drug information is wrong. After obtaining the drug classification information, OCR technology is used to extract the drug information data from the drug instructions image data. According to the drug information data of the instructions image, the drug classification information is determined, and then the drug classification result recognized by the image is compared with the drug classification information of the instructions image. If they are consistent, it is considered that we have correctly identified the classification information of the drug. If they are inconsistent, it is considered that the barcode recognition of the drug outer packaging is incorrect.

[0082] S5. Determine the drug classification result based on the comparison result of the drug classification information.

[0083] In this example, by obtaining the packaging barcode image data and instruction manual image data of the newly-entered drugs, image recognition is first performed on the packaging barcode image data to obtain the first drug classification information, and then text classification information recognition is performed on the instruction manual image data to obtain the second drug classification information. The first drug classification information is then compared with the second drug classification information. Based on the comparison result, the accurate drug classification result is determined. In the process of recognizing the drug packaging barcode data, the recognition of the drug instruction manual can be combined to assist in the correct judgment of the packaging barcode, thereby avoiding the technical problem of misrecognition of the barcode due to overexposure, which leads to incorrect drug classification.

[0084] In a possible implementation, performing image recognition on the package barcode image data to obtain the first drug classification information includes:

[0085] S201 , performing multi-angle feature fusion processing on the packaging barcode image data to obtain three-dimensional reconstructed barcode area image data.

[0086] S202: Perform image enhancement processing on the barcode region image data to obtain enhanced barcode region image data.

[0087] S203: Inputting the enhanced barcode region image data into a barcode corner point coordinate recognition model to obtain a barcode corner point coordinate recognition result.

[0088] S204: Determine the target barcode area according to the barcode corner point coordinate recognition result.

[0089] S205: Decode the target barcode area to obtain first drug classification information.

[0090] Multi-angle feature fusion and dynamic enhancement processing technologies address the problem of barcode feature loss caused by package reflections or deformation in traditional single-view image recognition. Three-dimensional reconstruction of multi-angle images integrates effective information from different viewing angles, significantly improving barcode integrity in reflective areas or on curved packaging. Image enhancement preprocessing enhances detail restoration in blurred or low-quality areas by optimizing local contrast and edge sharpening. This allows the subsequent corner location model to more accurately capture the barcode's geometric structural features, establishing a reliable image foundation for accurate decoding.

[0091] In this example, by establishing a closed-loop processing mechanism from feature fusion to dynamic repair, the identification information of pharmaceutical packaging can be stably extracted even in complex optical environments. The combination of the corner coordinate recognition model and the target area positioning can avoid misidentification problems caused by local overexposure or contamination. Even in highly reflective metal packaging scenarios, the true form of the barcode can be restored through a three-dimensional reconstruction compensation mechanism. This layered and progressive image processing logic not only improves the anti-interference ability of barcode recognition, but also provides more reliable data input for subsequent classification decisions through multi-stage feature optimization, ultimately ensuring the robustness and accuracy of the pharmaceutical classification system.

[0092] In a possible implementation, performing text classification information recognition on the instruction sheet image data to obtain the second drug classification information includes:

[0093] S301, extracting text information and coordinate information from the instruction manual image data;

[0094] S302, aligning the text information according to the coordinate information to obtain aligned text information;

[0095] S303, performing drug classification key field recognition on the aligned text information to obtain drug classification key field recognition results;

[0096] S304: Determine the second drug classification information according to the identification result of the drug classification key field.

[0097] Spatial alignment and semantic focus technologies effectively address text misalignment and information fragmentation caused by folding and wrinkling instructions. The system first extracts the text and its spatial coordinates, reconstructing the original layout structure through coordinate alignment to eliminate text distortion caused by camera angle or paper deformation. It then focuses on intelligent recognition of key classification fields, accurately locating core elements such as approval numbers and indications through contextual analysis, transforming discrete text information into structured semantic features. This dual spatial and semantic analysis mechanism ensures reliable extraction of classification criteria from complex layouts.

[0098] Furthermore, this example uses a hierarchical information processing process to improve the accuracy of key information extraction while ensuring text integrity. Text alignment restores the logical arrangement of the drug instructions, laying an accurate semantic foundation for subsequent field recognition; while the focused extraction of key fields eliminates redundant descriptions through a semantic filtering mechanism, directly targeting the core elements of classification decisions. This processing method allows accurate restoration of classification information through contextual reasoning and spatial correlation even in the case of partial defacement or blurred printing of the instructions, forming a dual verification with the barcode recognition of claim 2, significantly improving the system's adaptability to complex scenarios.

[0099] In a possible implementation, determining the second drug classification information according to the drug classification key field identification result includes:

[0100] S3041. Use a regional attention mechanism to perform field extraction on the key field recognition results of the drug classification to obtain the approval number, indication, and pharmacological classification description;

[0101] S3042. Match the drug-classification knowledge graph according to the approval number, indication, and pharmacological classification description to obtain second drug classification information.

[0102] Among them, through intelligent semantic focusing and knowledge association technology, a precise mapping mechanism from text fragments to classification decisions was established. Based on the aligned text in step S3, the regional attention mechanism dynamically analyzes the text layout and semantic weight, prioritizing fields with classification identification significance such as approval numbers and indications, and filtering out interfering descriptions through contextual association. The extracted core elements are then matched with the drug-classification knowledge graph in multiple dimensions, and the pre-constructed associations between drug attributes, pharmacological effects, and regulatory codes in the knowledge graph are used to verify the integrity and consistency of the classification logic.

[0103] In this example, a dual-verification architecture significantly improved the reliability of package insert parsing: an attention mechanism ensured the complete extraction of key classification features, preventing missed detections due to differences in text layout; while knowledge graph matching, through a structured medical knowledge system, transformed discrete text descriptions into verifiable basis for classification decisions. Even when the package insert contains abbreviations or non-standard expressions, the system was able to accurately associate the drug's standardized classification attributes through the semantic extension and reasoning capabilities of the knowledge graph. This complemented the barcode repair mechanism in step S2, creating a more fault-tolerant classification verification system.

[0104] In a possible implementation, determining a drug classification result based on the comparison result of the drug classification information includes:

[0105] S501. If the comparison result indicates that the drug classification information is consistent, determine that the drug classification result is the drug classification information determined jointly by the first drug classification information and the second drug classification information as the drug classification result;

[0106] S502: If the comparison result indicates that the drug classification information is inconsistent, an adaptive threshold segmentation algorithm is used to separate the package barcode image data to obtain separated barcode data.

[0107] S503, using a multi-frame dynamic compensation algorithm to repair the separated barcode data to obtain repaired barcode data;

[0108] S504: Decode the repaired barcode data to obtain third drug classification information;

[0109] S505: Compare the third drug classification information with the second drug classification information to obtain a secondary comparison result;

[0110] S506. If the secondary comparison result indicates that the drug classification information is consistent, confirm that the drug classification information jointly determined by the second drug classification information and the third drug classification information is the drug classification result.

[0111] A self-correction system for classification conflicts has been constructed through dynamic iterative repair and multiple verification mechanisms. When the initial barcode recognition result is inconsistent with the information in the instructions, the system first activates adaptive image separation technology to decouple features from overexposed or reflective areas, removing interference noise and restoring the core barcode structure. Subsequently, a multi-frame dynamic compensation algorithm, integrating valid code element features from historical acquisition frames, repairs the damaged barcode structure at the spatial and temporal levels, generating a third drug classification with higher integrity. A secondary comparison verification process is then used to automatically correct erroneous recognition results.

[0112] In this example, image restoration technology overcomes the physical limitations of single-shot recognition, dynamically optimizing the quality of the recognition input while preserving the original data features. A secondary comparison mechanism, leveraging the stability of the instructions, establishes a verification anchor point, combining the enhanced recognition capabilities of step S2 with the knowledge graph verification of step S4. Ultimately, even in scenarios where optical interference causes initial recognition failure, reliable classification results can still be output through iterative restoration and cross-validation. This fault-tolerant mechanism significantly improves the system's adaptability to extreme environments (such as strong reflections and damaged packaging), ensuring the ultimate accuracy of medical drug classification.

[0113] In line with the above, please see Figure 2 , Figure 2 The present invention provides a schematic diagram of the structure of a medicine classification device for a medical system. Figure 2 As shown, the device includes:

[0114] Acquisition unit 1, used to acquire the packaging barcode image data and instruction sheet image data of the newly-entered medicine;

[0115] A first processing unit 2 is configured to perform image recognition on the package barcode image data to obtain first drug classification information;

[0116] The second processing unit 3 is used to perform text classification information recognition on the instruction manual image data to obtain second drug classification information;

[0117] The third processing unit 4 is configured to compare the first drug classification information with the second drug classification information to obtain a drug classification information comparison result;

[0118] The determination unit 5 is configured to determine a drug classification result based on the comparison result of the drug classification information.

[0119] In a possible implementation, in the aspect of performing image recognition on the package barcode image data to obtain the first drug classification information, the first processing unit 2 is configured to:

[0120] Performing multi-angle feature fusion processing on the packaging barcode image data to obtain three-dimensional reconstructed barcode area image data;

[0121] Performing image enhancement processing on the barcode area image data to obtain enhanced barcode area image data;

[0122] Inputting the enhanced barcode area image data into a barcode corner point coordinate recognition model to obtain a barcode corner point coordinate recognition result;

[0123] Determining the target barcode area according to the barcode corner point coordinate recognition result;

[0124] The target barcode area is decoded to obtain first drug classification information.

[0125] In a possible implementation, in the aspect of performing text classification information recognition on the instruction sheet image data to obtain the second drug classification information, the second processing unit 3 is configured to:

[0126] extracting text information and coordinate information from the instruction manual image data;

[0127] aligning the text information according to the coordinate information to obtain aligned text information;

[0128] Performing drug classification key field recognition on the aligned text information to obtain drug classification key field recognition results;

[0129] The second drug classification information is determined based on the identification result of the drug classification key field.

[0130] In a possible implementation, in determining the second drug classification information according to the drug classification key field identification result, the second processing unit 3 is configured to:

[0131] The regional attention mechanism is used to extract the key fields of the drug classification recognition results to obtain the approval number, indication and pharmacological classification description;

[0132] According to the approval number, indication and pharmacological classification description, a drug-classification knowledge graph is used for matching to obtain the second drug classification information.

[0133] In a possible implementation, in determining the drug classification result based on the drug classification information comparison result, the determining unit 5 is configured to:

[0134] If the information indicated by the comparison result is that the drug classification information is consistent, determining that the drug classification result is the drug classification information determined jointly by the first drug classification information and the second drug classification information as the drug classification result;

[0135] If the comparison result indicates that the drug classification information is inconsistent, an adaptive threshold segmentation algorithm is used to separate the package barcode image data to obtain separated barcode data;

[0136] Using a multi-frame dynamic compensation algorithm to repair the separated barcode data to obtain repaired barcode data;

[0137] Decoding the repaired barcode data to obtain third drug classification information;

[0138] Comparing the third drug classification information with the second drug classification information to obtain a secondary comparison result;

[0139] If the secondary comparison result indicates that the drug classification information is consistent, the drug classification information jointly determined by the second drug classification information and the third drug classification information is confirmed as the drug classification result.

[0140] For the same example as above, please refer to Figure 3 , Figure 3 A schematic structural diagram of a terminal provided in an embodiment of the present application, as shown in the figure, includes a processor, an input device, an output device, and a memory, the processor, the input device, the output device, and the memory being interconnected, wherein the memory is used to store a computer program, the computer program including program instructions, the processor being configured to call the program instructions, and the program including instructions for executing the following steps;

[0141] Obtain the packaging barcode image data and instruction sheet image data of newly-entered drugs;

[0142] Performing image recognition on the package barcode image data to obtain first drug classification information;

[0143] Performing text classification information recognition on the instruction manual image data to obtain second drug classification information;

[0144] Comparing the first drug classification information with the second drug classification information to obtain a drug classification information comparison result;

[0145] The drug classification result is determined based on the comparison result of the drug classification information.

[0146] In this example, by establishing a closed-loop processing mechanism from feature fusion to dynamic repair, the identification information of pharmaceutical packaging can be stably extracted even in complex optical environments. The combination of the corner coordinate recognition model and the target area positioning can avoid misidentification problems caused by local overexposure or contamination. Even in highly reflective metal packaging scenarios, the true form of the barcode can be restored through a three-dimensional reconstruction compensation mechanism. This layered and progressive image processing logic not only improves the anti-interference ability of barcode recognition, but also provides more reliable data input for subsequent classification decisions through multi-stage feature optimization, ultimately ensuring the robustness and accuracy of the pharmaceutical classification system.

[0147] The above mainly introduces the scheme of the embodiment of the present application from the perspective of the execution process on the method side. It is understandable that, in order to implement the above functions, the terminal includes hardware structures and / or software modules corresponding to the execution of each function. Those skilled in the art should easily realize that, in combination with the units and algorithm steps of the various examples described in the embodiments provided herein, the present application can be implemented in the form of hardware or a combination of hardware and computer software. Whether a function is executed in a hardware or computer software driven hardware manner depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0148] The embodiment of the present application can divide the terminal into functional units according to the above method example. For example, each functional unit can be divided according to each function, or two or more functions can be integrated into one processing unit. The above integrated unit can be implemented in the form of hardware or in the form of software functional units. It should be noted that the division of units in the embodiment of the present application is schematic and is only a logical functional division. In actual implementation, there may be other division methods.

[0149] An embodiment of the present application also provides a computer storage medium, wherein the computer storage medium stores a computer program for electronic data exchange, and the computer program enables a computer to execute part or all of the steps of any medical system drug classification method described in the above method embodiments.

[0150] An embodiment of the present application also provides a computer program product, which includes a non-transitory computer-readable storage medium storing a computer program, and the computer program enables a computer to execute part or all of the steps of any medical system drug classification method recorded in the above method embodiments.

[0151] It should be noted that for the aforementioned method embodiments, for the sake of simplicity, they are all expressed as a series of action combinations, but those skilled in the art should be aware that this application is not limited by the order of the actions described, because according to this application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by this application.

[0152] In the above embodiments, the description of each embodiment has its own focus. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0153] In the several embodiments provided in this application, it should be understood that the disclosed devices can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the units is merely a logical function division. 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. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, and the indirect coupling or communication connection of devices or units can be electrical or other forms.

[0154] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0155] In addition, the functional units in the various embodiments of the application may be integrated into a single processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software program modules.

[0156] If the integrated unit is implemented in the form of a software program module and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a memory, including a number of instructions for enabling a computer device (which can be a personal computer, a server or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned memory includes: various media that can store program codes, such as a U disk, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk.

[0157] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable memory, which may include: a flash drive, a read-only memory, a random access memory, a magnetic disk or an optical disk, etc.

[0158] The above is a detailed introduction to the embodiments of the present application. Specific examples are used herein to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method and core idea of ​​the present application. At the same time, for those skilled in the art, according to the idea of ​​the present application, there may be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation on the present application.

Claims

1. A drug classification method for a medical system, characterized in that: include: Obtain the packaging barcode image data and instruction sheet image data of newly-entered drugs; Performing image recognition on the package barcode image data to obtain first drug classification information; Performing text classification information recognition on the instruction manual image data to obtain second drug classification information; Comparing the first drug classification information with the second drug classification information to obtain a drug classification information comparison result; The drug classification result is determined based on the comparison result of the drug classification information.

2. The drug classification method of the medical system according to claim 1, characterized in that: The performing image recognition on the package barcode image data to obtain first drug classification information includes: Performing multi-angle feature fusion processing on the packaging barcode image data to obtain three-dimensional reconstructed barcode area image data; Performing image enhancement processing on the barcode area image data to obtain enhanced barcode area image data; Inputting the enhanced barcode area image data into a barcode corner point coordinate recognition model to obtain a barcode corner point coordinate recognition result; Determining the target barcode area according to the barcode corner point coordinate recognition result; The target barcode area is decoded to obtain first drug classification information.

3. The drug classification method of the medical system according to claim 1, characterized in that: The step of performing text classification information recognition on the instruction manual image data to obtain second drug classification information includes: extracting text information and coordinate information from the instruction manual image data; aligning the text information according to the coordinate information to obtain aligned text information; Performing drug classification key field recognition on the aligned text information to obtain drug classification key field recognition results; The second drug classification information is determined based on the identification result of the drug classification key field.

4. The drug classification method of the medical system according to claim 3, characterized in that: The determining of the second drug classification information according to the drug classification key field identification result includes: The regional attention mechanism is used to extract the key fields of the drug classification recognition results to obtain the approval number, indication and pharmacological classification description; According to the approval number, indication and pharmacological classification description, a drug-classification knowledge graph is used for matching to obtain the second drug classification information.

5. The drug classification method of a medical system according to claim 1, characterized in that: Determining the drug classification result based on the comparison result of the drug classification information includes: If the information indicated by the comparison result is that the drug classification information is consistent, determining that the drug classification result is the drug classification information determined jointly by the first drug classification information and the second drug classification information as the drug classification result; If the comparison result indicates that the drug classification information is inconsistent, an adaptive threshold segmentation algorithm is used to separate the package barcode image data to obtain separated barcode data; Using a multi-frame dynamic compensation algorithm to repair the separated barcode data to obtain repaired barcode data; Decoding the repaired barcode data to obtain third drug classification information; Comparing the third drug classification information with the second drug classification information to obtain a secondary comparison result; If the secondary comparison result indicates that the drug classification information is consistent, the drug classification information jointly determined by the second drug classification information and the third drug classification information is confirmed as the drug classification result.

6. A medicine classification device for a medical system, characterized in that: The device comprises: An acquisition unit, used to acquire the package barcode image data and instruction sheet image data of newly-entered drugs; a first processing unit, configured to perform image recognition on the package barcode image data to obtain first drug classification information; a second processing unit, configured to perform text classification information recognition on the instruction manual image data to obtain second drug classification information; a third processing unit, configured to compare the first drug classification information with the second drug classification information to obtain a drug classification information comparison result; The determination unit is used to determine the drug classification result according to the comparison result of the drug classification information.

7. The medicine classification device of the medical system according to claim 6, characterized in that: In the aspect of performing image recognition on the package barcode image data to obtain the first drug classification information, the first processing unit is configured to: Performing multi-angle feature fusion processing on the packaging barcode image data to obtain three-dimensional reconstructed barcode area image data; Performing image enhancement processing on the barcode area image data to obtain enhanced barcode area image data; Inputting the enhanced barcode area image data into a barcode corner point coordinate recognition model to obtain a barcode corner point coordinate recognition result; Determining the target barcode area according to the barcode corner point coordinate recognition result; The target barcode area is decoded to obtain first drug classification information.

8. The medicine classification device of the medical system according to claim 6, characterized in that: In the aspect of determining the drug classification result based on the comparison result of the drug classification information, the determining unit is configured to: If the information indicated by the comparison result is that the drug classification information is consistent, determining that the drug classification result is the drug classification information determined jointly by the first drug classification information and the second drug classification information as the drug classification result; If the comparison result indicates that the drug classification information is inconsistent, an adaptive threshold segmentation algorithm is used to separate the package barcode image data to obtain separated barcode data; Using a multi-frame dynamic compensation algorithm to repair the separated barcode data to obtain repaired barcode data; Decoding the repaired barcode data to obtain third drug classification information; Comparing the third drug classification information with the second drug classification information to obtain a secondary comparison result; If the secondary comparison result indicates that the drug classification information is consistent, the drug classification information jointly determined by the second drug classification information and the third drug classification information is confirmed as the drug classification result.

9. A terminal, characterized in that: The method comprises a processor, an input device, an output device and a memory, wherein the processor, the input device, the output device and the memory are interconnected, wherein the memory is used to store a computer program, the computer program includes program instructions, and the processor is configured to call the program instructions to execute the drug classification method of the medical system according to any one of claims 1 to 5.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, wherein the computer program includes program instructions. When the program instructions are executed by a processor, the processor executes the drug classification method of the medical system according to any one of claims 1 to 5.

Images