Electric power material warehouse-in and warehouse-out management method and system based on image recognition, medium and processor

Abstract

The invention discloses an electric power material warehouse-in and warehouse-out management method and system based on image recognition, a medium and a processor, and relates to the technical field of warehouse management, and the method comprises the steps: obtaining the basic information of electric power materials, and dividing the electric power materials into large-class materials and subdivided materials according to the basic information; acquiring image data of electric power materials to obtain an electric power material image; processing the electric power material image through an image difference method, extracting the contour area of each electric power material, and calculating an area threshold according to the contour area; extracting text information in the to-be-detected electric power material image through an optical character recognition model, and matching the text information with a database to obtain a recognition result of a large class of materials; and performing category identification on the to-be-detected electric power material image, and outputting an electric power material identification result after contour area threshold judgment. Different warehouse-in and warehouse-out management processes can be adopted for different types of electric power materials, the management efficiency is improved, and the warehouse-in and warehouse-out management data accuracy is enhanced.

Description

Technical Field

[0001] The present invention relates to the technical field of warehouse management, and in particular to a method, system, medium and processor for managing the entry and exit of electric power materials based on image recognition. Background Art

[0002] With the acceleration of the digital transformation of the power industry, warehouse management will be integrated with advanced technologies such as the Internet of Things and blockchain to form a more intelligent and efficient power material management system. This will not only improve the overall level of power material management, but also provide strong support for the sustainable development of the power industry.

[0003] In the current power storage environment, power materials are not only different in categories, but also in models. Different models are suitable for different scenarios, especially in power repair scenarios, which involve the entry and exit of many spare parts and emergency repair parts. Spare parts and emergency repair parts of different models correspond to a single use condition. When storing, the warehouse also stores them separately according to the corresponding model, rather than managing the same type of materials together. The original management method of the power material warehouse was to equip material record cards according to the material model, and to manage the entry and exit of materials of different models separately. Therefore, for materials with multiple models, it is also necessary to manage the entry and exit of materials according to different material models. The warehouse material entry and exit management uses "material record cards" to record the entry and exit of materials and the balance within the cycle, and then the warehouse manager organizes the material cards and takes them to the material system. In these processes, whenever someone picks up materials, the time and quantity of the materials are manually recorded on the record card. Due to the large variety and quantity of materials, it is easy to cause problems such as confusion on the card and unclear handwriting, and it requires a lot of manpower, material resources and time.

[0004] In view of this, a method, system, medium and processor for managing the entry and exit of electric power materials based on image recognition are needed. Summary of the invention

[0005] In view of the problem that the current power material warehouse management methods in the prior art are insufficient, the present invention provides a power material in-and-out warehouse management method, system, medium and processor based on image recognition, which can adopt different in-and-out warehouse management processes for different types of power materials, improve management efficiency and enhance the accuracy of in-and-out warehouse management data. The specific technical solution is as follows:

[0006] A method for managing the entry and exit of electric power materials based on image recognition, comprising:

[0007] Obtain basic information of power materials, classify power materials into major materials and sub-materials according to the basic information, and store them in the database;

[0008] Acquire image data of electric power materials to obtain images of electric power materials;

[0009] Processing the power material image by an image difference method, extracting the contour area of ​​each power material, calculating the area threshold according to the contour area, and storing it in a database;

[0010] The optical character recognition model is used to extract text information from the image of the power material to be tested, and the text information is matched with the database to obtain the recognition results of the major categories of materials;

[0011] The image of the electric power material to be tested is subjected to category recognition. If the electric power material is a subdivided material, the electric power material recognition result is output after judging through the contour area threshold, otherwise the electric power material recognition result is directly output.

[0012] Preferably, the acquiring of image data of electric power materials to obtain an image of electric power materials comprises:

[0013] Image data of each electric power material in different forms, positions and quantities are obtained to obtain images of the electric power material.

[0014] Preferably, the processing of the electric power material image by an image difference method, extracting the contour area of ​​each electric power material, and obtaining the area threshold according to the contour area includes:

[0015] The power material image is Gaussian blurred and then subjected to a difference operation with the background image to obtain a difference image;

[0016] The difference image is input into the HSV color model and converted into an HSV channel, and a brightness image is extracted;

[0017] Binarizing the brightness image to obtain a binary image;

[0018] Obtaining the outline of the power material according to the binary image, and calculating the outline area of ​​the power material;

[0019] The area threshold of the power material is set according to the outline area of ​​the power material.

[0020] Preferably, a method for managing the entry and exit of electric power materials based on image recognition further includes:

[0021] A data enhanced image of electric power materials is generated for the binary image through an image ROI.

[0022] Preferably, generating a data enhanced image of the electric power material through an image ROI for the electric power material image comprises:

[0023] Binarizing the power material image to obtain a binary image;

[0024] Obtaining the minimum circumscribed rectangle coordinates of the electric power material according to the binary image;

[0025] Segmenting power material pixels from the power material image according to the minimum circumscribed rectangle coordinates;

[0026] The power material pixels are pasted into the original power material image at different angles and positions to obtain a data enhanced image of the power material.

[0027] Preferably, the extracting text information from the image of the power material to be tested by the optical character recognition model and matching the text information with the database to obtain the recognition results of the major categories of materials includes:

[0028] Extract text information from the image of the power material to be tested through an optical character recognition model;

[0029] Performing word segmentation processing on the text information;

[0030] By traversing the segmented text one by one and comparing it with the text in the database, the matching rate results of major categories of materials are obtained, and the major category of material model with the highest matching degree is selected as the identified one.

[0031] Preferably, the performing of category recognition on the image of the electric power material to be tested and outputting the electric power material recognition result after judging by the contour area threshold comprises:

[0032] If the electric power material is determined to be a subdivided material, the contour area of ​​the image of the electric power material to be tested is calculated;

[0033] Compare the contour area of ​​the image of the electric power material to be tested with the area thresholds of all models of electric power materials in the database to obtain an area difference;

[0034] If the area difference is within a preset threshold, the corresponding electric power material model is determined.

[0035] An electric power material in-and-out storage management system based on image recognition, applied to the electric power material in-and-out storage management method based on image recognition according to any one of claims 1 to 7, comprising:

[0036] The information and image acquisition unit is used to acquire basic information of the power materials, and classify the power materials into major materials and sub-materials according to the basic information, and store them in the database; acquire image data of the power materials and obtain images of the power materials;

[0037] An image processing unit, used to process the power material image by an image difference method, extract the contour area of ​​each power material, calculate the area threshold according to the contour area, and store it in a database;

[0038] The material analysis and identification unit is used to extract text information from the image of the power material to be tested through the optical character recognition model, and match the text information with the database to obtain the identification results of the major categories of materials; perform category identification on the image of the power material to be tested, and output the power material identification results after judging through the contour area threshold.

[0039] A computer-readable storage medium includes a stored program, wherein when the program is executed, the device where the computer-readable storage medium is located is controlled to execute the aforementioned method for managing the entry and exit of electric power materials based on image recognition.

[0040] A processor is used to run a program, wherein the program executes the aforementioned method for managing the entry and exit of electric power materials based on image recognition when running.

[0041] Compared with the prior art, the present invention has the following beneficial effects:

[0042] The invention discloses an in-and-out storage management method for electric power materials based on image recognition, which obtains basic information of electric power materials, divides the electric power materials into major materials and subdivided materials according to the basic information, and stores them in a database; obtains image data of electric power materials to obtain images of electric power materials; processes the images of electric power materials by image difference method, extracts contour areas of each electric power material, calculates area thresholds according to contour areas, and stores them in a database; extracts text information in the images of electric power materials to be tested by an optical character recognition model, and matches the text information with the database to obtain identification results of major materials; performs category identification on the images of electric power materials to be tested, and outputs the identification results of electric power materials after judging by contour area thresholds. The invention can adopt different in-and-out storage management processes for different types of electric power materials, improve management efficiency, and enhance the accuracy of in-and-out storage management data. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following is a brief introduction to the drawings required for the specific embodiments or the description of the prior art. In all the drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, the elements or parts are not necessarily drawn according to the actual scale.

[0044] Figure 1 The present invention is a flow chart of a method for managing the entry and exit of electric power materials based on image recognition.

[0045] Figure 2 The present invention is a flowchart of another embodiment of a method for managing the entry and exit of electric power materials based on image recognition.

[0046] Figure 3The present invention is a flowchart of another embodiment of a method for managing the entry and exit of electric power materials based on image recognition.

[0047] Figure 4 The present invention is a flowchart of another embodiment of a method for managing the entry and exit of electric power materials based on image recognition.

[0048] Figure 5 This is a schematic diagram of an electric power material in-and-out storage management system based on image recognition according to the present invention. DETAILED DESCRIPTION

[0049] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0050] It should be understood that when used in this specification and the appended claims, the terms "include" and "comprises" indicate the presence of described features, integers, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or combinations thereof.

[0051] It should also be understood that the terms used in the present specification are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the present specification and the appended claims, the singular forms "a", "an" and "the" are intended to include plural forms unless the context clearly indicates otherwise.

[0052] It should be further understood that the term "and / or" used in the present description and the appended claims refers to and includes any and all possible combinations of one or more of the associated listed items.

[0053] See the following examples Figures 1 to 5 .

[0054] The present application provides a method for managing the entry and exit of electric power materials based on image recognition, including:

[0055] Step S01, obtaining basic information of electric power materials, classifying the electric power materials into major materials and sub-material materials according to the basic information, and storing them in a database;

[0056] In this embodiment, basic information includes category, model, key information and important information, etc., and then can be divided into major categories of materials and subdivided materials including type, model and other key words according to the information. Major categories of materials mainly refer to whole boxes or a general category, which can be directly identified according to the text information of power materials; subdivided categories are some scattered power materials, which may include a lot, which cannot be directly identified according to text information, but can only be identified according to image data features and shape differences. Storing this information of power materials in the database can facilitate subsequent comparison and matching.

[0057] Step S02, acquiring image data of electric power materials to obtain an image of electric power materials;

[0058] In this embodiment, by acquiring image data of each electric power material in different forms, different positions and different quantities, the electric power material image is obtained. The recognition accuracy of the subdivided materials can be improved through different forms and positions, so that accurate recognition can be achieved regardless of how the subsequent electric power materials are placed in and out of the warehouse.

[0059] Since power materials are divided into major materials and sub-materials for identification, obtaining images of power materials is an important step. The major materials can be identified through text information recognition of power material images, and sub-materials can be identified through shape recognition of power material images. By storing power material images in a database, subsequent comparison and matching can be facilitated.

[0060] Step S03, processing the electric power material image by an image difference method, extracting the contour area of ​​each electric power material, calculating the area threshold according to the contour area, and storing it in a database;

[0061] The image difference method is to subtract the pixel values ​​of the target image from the background image to weaken the similar parts of the two images and highlight the changed parts of the images. Since the background is fixed in the actual recognition scene, the pixel difference between the background image and the image to be recognized can be calculated to highlight the pixels of the material to be recognized. Therefore, the contour feature information of the power material can be obtained through the image difference method.

[0062] Step S04: extracting text information from the image of the power material to be tested through an optical character recognition model, and matching the text information with the database to obtain recognition results of major categories of materials;

[0063] Since the product name, product model, company name and other information printed in the whole box packaging of different power materials are different, the information printed on the outside of the power material box can be used as a unique identifier, and the CTPN and CRNN models are used to locate the text position in the power material image and extract the text information. The text information is segmented and stored in the database, and the identifier of the power material box is used to represent the unique material box. When a power material box enters and leaves the warehouse, the unique identifier of the power material box to be identified can be used to perform text matching with the identifier in the database, and the matching rate is used as the basis for returning the result, and the one with the highest matching rate is selected as the final recognition result, so that the corresponding recognition result of the major category of materials can be output.

[0064] In this embodiment, in order to ensure the accuracy of recognition, it is necessary to use the model to extract text information from the power material box in the actual entry and exit scenario and save it in the database after word segmentation when constructing the database, rather than manually identifying the identifier and storing it in the database. In this way, in the processing of small model text that is difficult to distinguish by the model, the model standards can be used for judgment before and after, and when the power material box to be identified is entering and leaving the warehouse, the matching rate of the corresponding material category can be maximized.

[0065] The formula for calculating the matching rate is as follows:

[0066]

[0067] Where Prob is the matching rate; N match The number of successful matching words for the power material identifier in the database; N total is the total number of power material identifiers in the database.

[0068] Step S05: classify the power material image to be tested, and output the power material identification result after judging by the contour area threshold.

[0069] If the electric power material is determined to be a subdivided material, the contour area of ​​the image of the electric power material to be tested is calculated;

[0070] Compare the contour area of ​​the image of the electric power material to be tested with the area thresholds of all models of electric power materials in the database to obtain an area difference;

[0071] If the area difference is within a preset threshold, the corresponding electric power material model is determined.

[0072] By setting the threshold of the area difference, the threshold can be adjusted according to actual experience to achieve accurate recognition.

[0073] The method for managing the entry and exit of electric power materials based on image recognition of the present invention can improve management efficiency and enhance the accuracy of entry and exit management data by adopting different entry and exit management processes for different types of electric power materials.

[0074] Specifically, in a preferred implementation of the present application, the image of the power material is processed by an image difference method to extract the contour area of ​​each power material, and the area threshold is obtained according to the contour area, including:

[0075] Step S31, performing a difference operation between the power material image and the background image after Gaussian blurring to obtain a difference image;

[0076] Gaussian blur is used to reduce image noise and detail levels. By performing Gaussian blur on the power material image, some unnecessary details and noise can be removed, making subsequent processing easier. The difference operation performs a difference operation on the Gaussian blurred image and the background image (usually a known or preset image without power materials). The result of the difference operation is a new image in which the power material part will appear prominent and the background part will be suppressed or eliminated. This difference image will be used for further processing.

[0077] Step S32, converting the difference image into an HSV color model and extracting a brightness image;

[0078] HSV color model, where H stands for hue, S stands for saturation, and V stands for value. Compared with the RGB model, the HSV model is more intuitive in color processing. Convert the difference map from the RGB color space to the HSV color space. In the HSV model, the V channel represents the brightness information of the image. By extracting the V channel (i.e., the brightness map), we can get an image containing only brightness information, which will help the subsequent binarization process.

[0079] Step S33, binarizing the brightness map to obtain a binary image;

[0080] Convert the pixel values ​​of the brightness image to 0 or 255 (or other values ​​representing black and white), that is, create an image containing only black and white. In this process, a threshold is usually set, and pixels with brightness above the threshold are set to white (or black), and pixels below the threshold are set to black (or white). In this way, the power material part will be clearly presented in the form of white (or black).

[0081] Step S34, obtaining the outline of the power material according to the binary image, and calculating the outline area of ​​the power material;

[0082] The contour of the power material is extracted from the binary image, and the area enclosed by the extracted contour is calculated. This area represents the size of the power material in the image.

[0083] Step S35: setting an area threshold of the electric power material according to the outline area of ​​the electric power material.

[0084] The same type of power material can identify the contour features of multiple power materials, and then the contour area can be obtained according to the contour features. The area threshold can be obtained by calculating the average value of the contour areas of multiple power materials of the same type.

[0085] The area threshold is calculated as follows:

[0086]

[0087] In the formula, S t is the area threshold of a certain type of electric power material; S1 represents the contour area of ​​the nth identified electric power material of this type.

[0088] Specifically, in a preferred embodiment of the present application, a method for managing the entry and exit of electric power materials based on image recognition further includes:

[0089] A data enhanced image of electric power materials is generated for the binary image through an image ROI.

[0090] In a specific implementation, generating a data enhanced image of the electric power material through the image ROI includes:

[0091] Step S301, binarizing the power material image to obtain a binary image;

[0092] Step S302, obtaining the minimum circumscribed rectangle coordinates of the electric power material according to the binary image;

[0093] After determining the position and range of the power material in the image, the boundary points of the power material (i.e., the white area) are found by traversing the pixels in the binary image, and then the minimum rectangle that can completely surround the power material is calculated based on these boundary points, and the coordinate information of this rectangle is recorded.

[0094] Step S303, segmenting the power material pixels from the power material image according to the minimum circumscribed rectangle coordinates;

[0095] The image portion of the electric power material is accurately extracted from the original image, and the corresponding area is cut out from the original electric power material image using the minimum bounding rectangle coordinates obtained in step S302 as a reference. This area contains the pixels of the electric power material.

[0096] Step S304: Paste the power material pixels into the original power material image at different angles and positions to obtain a data enhanced image of the power material.

[0097] The electric power material pixel blocks segmented in step S303 are rotated, and different rotation angles (such as 90 degrees, 180 degrees, etc.) or random angles can be selected; the rotated electric power material pixel blocks are moved to different positions in the image, which can be random or according to certain rules; the processed electric power material pixel blocks are pasted back into the original image to generate new image data. In this process, it may be necessary to consider the processing of image boundaries, such as cropping or filling when exceeding the boundaries. In this way, multiple electric power material images of different angles and positions can be generated, effectively increasing the scale of the data set and improving the generalization ability of model training.

[0098] In this embodiment, the background and material pixels are segmented by the ROI data enhancement method, and the segmented power material images are pasted on the background image according to different positions and postures. This not only ensures the authenticity of the power material images, but also can more widely reflect the real power material in-and-out storage scenes, while reducing the complexity of data collection and improving the generalization ability of the model.

[0099] Specifically, in a preferred embodiment of the present application, the text information in the image of the power material to be tested is extracted by the optical character recognition model, and the text information is matched with the database to obtain the recognition results of the major categories of materials, including:

[0100] Step S41, extracting text information from the image of the power material to be tested by using an optical character recognition model;

[0101] Automatically extract text information from images of power materials using optical character recognition (OCR) technology. OCR technology can recognize text in images and convert it into a computer-readable text format. In power material management, this usually means reading key information such as model, serial number, manufacturer, etc. from material labels, packaging, or any image containing material information.

[0102] Step S42, performing word segmentation processing on the text information;

[0103] Through word segmentation, long text can be split into smaller, meaningful units (i.e. words) for easy subsequent processing.

[0104] Step S43: By traversing the segmented texts one by one and comparing and matching them with the texts in the database, the result of the major material matching rate is obtained, and the major material model with the highest matching degree is selected as the identified major material model.

[0105] The segmented text is compared and matched with the power material information stored in the database. The database usually contains detailed information on various power materials, including models, specifications, manufacturers, etc. The similarity or matching degree between the texts is calculated to determine the major types and models of the materials to be tested. The result with the highest matching degree is usually considered to be the correct recognition result.

[0106] In this embodiment, the efficiency and accuracy of material management are greatly improved by automatically identifying and classifying power materials. It not only reduces manual intervention and reduces the risk of human error, but also speeds up material processing, allowing power companies to manage their material resources more efficiently.

[0107] The present application also provides an electric power material in-and-out storage management system based on image recognition, including:

[0108] The information and image acquisition unit is used to acquire basic information of the power materials, and classify the power materials into major materials and sub-materials according to the basic information, and store them in the database; acquire image data of the power materials and obtain images of the power materials;

[0109] An image processing unit, used to process the power material image by an image difference method, extract the contour area of ​​each power material, calculate the area threshold according to the contour area, and store it in a database;

[0110] The material analysis and identification unit is used to extract text information from the image of the power material to be tested through an optical character recognition model, and match the text information with the database to obtain a matching result; if the matching result is higher than the matching threshold, the power material identification result is output; if the text matching result is lower than the matching threshold, the power material image is identified in categories. If the power material is a subdivided material, the power material identification result is output after judging through the contour area threshold, otherwise the power material identification result is directly output.

[0111] The functional explanation of each unit in this embodiment is the same as the technical effect of an embodiment of a method for managing the entry and exit of electric power materials based on image recognition, and will not be repeated here.

[0112] The present application also provides a computer-readable storage medium, which includes a stored program, wherein when the program is running, the device where the computer-readable storage medium is located is controlled to execute the aforementioned power material in-and-out storage management method based on image recognition.

[0113] The technical effect of this embodiment is the same as the technical effect of an embodiment of a method for managing the entry and exit of electric power materials based on image recognition, and will not be repeated here.

[0114] The present invention can be used in many general or special computer system environments or configurations, such as personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronic devices, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, and the like.

[0115] The present application also provides a processor, which is used to run a program, wherein the program executes the aforementioned power material in-and-out warehouse management method based on image recognition when running.

[0116] The technical effect of this embodiment is the same as the technical effect of an embodiment of a method for managing the entry and exit of electric power materials based on image recognition, and will not be repeated here.

[0117] The processor in this embodiment may be a central processing unit (CPU), a controller, a microcontroller, or other data processing chips.

[0118] Those of ordinary skill in the art will appreciate that the units of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition of each example has been generally described in terms of function in the above description. Whether these functions are performed in hardware or software 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 the present invention.

[0119] In the embodiments provided by the present invention, it should be understood that the division of units is only a logical function division, and there may be other division methods in actual implementation, for example, multiple units can be combined into one unit, one unit can be split into multiple units, or some features can be ignored, etc.

[0120] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0121] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, 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, which is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-0nlyMemory), random access memory (RAM, RandomAccessMemory), mobile hard disk, magnetic disk or optical disk, etc., which can store program code.

[0122] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein by equivalents. These modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be included in the scope of the claims and specification of the present invention.

Claims

1. A method for managing the entry and exit of electric power materials based on image recognition, characterized in that: include: Obtain basic information of power materials, classify power materials into major materials and sub-materials according to the basic information, and store them in the database; Acquire image data of electric power materials to obtain images of electric power materials; Processing the power material image by an image difference method, extracting the contour area of ​​each power material, calculating the area threshold according to the contour area, and storing it in a database; The optical character recognition model is used to extract text information from the image of the power material to be tested, and the text information is matched with the database to obtain the recognition results of the major categories of materials; The image of the electric power material to be tested is identified by category, and the electric power material identification result is output after judging by the contour area threshold.

2. The method for managing the entry and exit of electric power materials based on image recognition according to claim 1 is characterized in that: The acquiring of the image data of the electric power material to obtain the image of the electric power material comprises: Image data of each electric power material in different forms, positions and quantities are obtained to obtain images of the electric power material.

3. The method for managing the entry and exit of electric power materials based on image recognition according to claim 2 is characterized in that: The processing of the electric power material image by the image difference method, extracting the contour area of ​​each electric power material, and obtaining the area threshold according to the contour area includes: The power material image is Gaussian blurred and then subjected to a difference operation with the background image to obtain a difference image; The difference image is input into the HSV color model and converted into an HSV channel, and a brightness image is extracted; Binarizing the brightness image to obtain a binary image; Obtaining the outline of the power material according to the binary image, and calculating the outline area of ​​the power material; The area threshold of the power material is set according to the outline area of ​​the power material.

4. The method for managing the entry and exit of electric power materials based on image recognition according to claim 3 is characterized in that: Also includes: A data enhanced image of electric power materials is generated for the binary image through an image ROI.

5. The method for managing the entry and exit of electric power materials based on image recognition according to claim 4 is characterized in that: The generating a data enhancement image of the electric power material through the image ROI of the electric power material image comprises: Binarizing the power material image to obtain a binary image; Obtaining the minimum circumscribed rectangle coordinates of the electric power material according to the binary image; Segmenting power material pixels from the power material image according to the minimum circumscribed rectangle coordinates; The power material pixels are pasted into the original power material image at different angles and positions to obtain a data enhanced image of the power material.

6. The method for managing the entry and exit of electric power materials based on image recognition according to claim 1, characterized in that: The optical character recognition model is used to extract text information from the image of the power material to be tested, and the text information is matched with the database to obtain the recognition results of the major categories of materials, including: Extract text information from the image of the power material to be tested through an optical character recognition model; Performing word segmentation processing on the text information; By traversing the segmented text one by one and comparing it with the text in the database, the matching rate results of major categories of materials are obtained, and the major category of material model with the highest matching degree is selected as the identified one.

7. The method for managing the entry and exit of electric power materials based on image recognition according to claim 6 is characterized in that: The method of performing category recognition on the image of the power material to be tested and outputting the power material recognition result after judging by the contour area threshold comprises: If the power material is judged to be a subdivided material, calculate the contour area of ​​the image of the power material to be tested Compare the contour area of ​​the image of the electric power material to be tested with the area thresholds of all models of electric power materials in the database to obtain an area difference; If the area difference is within a preset threshold, the corresponding electric power material model is determined.

8. An electric power material in-and-out storage management system based on image recognition, characterized in that: The method for managing the entry and exit of electric power materials based on image recognition as described in any one of claims 1 to 7 comprises: The information and image acquisition unit is used to acquire basic information of the power materials, and classify the power materials into major materials and sub-materials according to the basic information, and store them in the database; acquire image data of the power materials and obtain images of the power materials; An image processing unit, used to process the power material image by an image difference method, extract the contour area of ​​each power material, calculate the area threshold according to the contour area, and store it in a database; The material analysis and identification unit is used to extract text information from the image of the power material to be tested through the optical character recognition model, and match the text information with the database to obtain the identification results of the major categories of materials; perform category identification on the image of the power material to be tested, and output the power material identification results after judging through the contour area threshold.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored program, wherein when the program is executed, the device where the computer-readable storage medium is located is controlled to execute the method for managing the entry and exit of electric power materials based on image recognition as described in any one of claims 1 to 7.

10. A processor, characterized in that: The processor is used to run a program, wherein the program, when running, executes the method for managing the entry and exit of electric power materials based on image recognition as described in any one of claims 1 to 7.

Images