CN-121982730-A - Intelligent identification method, device, equipment and storage medium for terminal strip information

Abstract

The application relates to the technical field of image processing, in particular to an intelligent identification method and device for terminal strip information, comprising the steps of acquiring a terminal strip image to be processed; the method comprises the steps of carrying out text recognition on a terminal strip image to obtain terminal strip text information, carrying out target mask segmentation on the terminal strip image, outputting position coordinates corresponding to the terminal strip text information, establishing a corresponding matching relation between terminal serial numbers and wiring contents, combining a position area corresponding to the terminal strip name, carrying out verification optimization on the recognized terminal strip name to obtain an optimized target terminal strip name, integrating the corresponding matching relation between the target terminal strip name, the terminal serial numbers and the wiring contents, outputting terminal strip information, and comparing the terminal strip information with terminal strip design data to obtain difference information. Therefore, the problems of low information identification efficiency, inaccurate position matching and verification dependence on manual work of the traditional terminal strip are solved through cooperative fusion of OCR and target detection and automatic comparison of the OCR and the design data.

Inventors

• CAO ZHAOYANG
• YANG JIA
• Chai Baozheng
• GUO HUI
• FENG JUN
• HE XIAOYU
• JIANG HAN
• LI JIN
• YU WEIGUO
• MENG WEI
• JIANG NENGREN
• YUAN XIAOPING
• XIE SHICHUN
• ZHANG XIN
• WEI ZHENFENG

Assignees

• 长园深瑞继保自动化有限公司

Dates

Publication Date

20260505

Application Date

20251219

Claims (11)

1. 1. An intelligent recognition method for terminal strip information is characterized by comprising the following steps: acquiring a terminal strip image to be processed; text recognition is carried out on the terminal strip image to obtain terminal strip text information, wherein the terminal strip text information at least comprises a terminal strip name, a terminal sequence number and wiring content; performing target mask segmentation on the terminal strip image, and outputting position coordinates corresponding to the terminal strip text information, wherein the position coordinates are represented by vertex coordinates of a rectangular frame and at least comprise coordinate information of the upper left corner and the lower right corner of the rectangular frame; based on the position coordinates, establishing a corresponding matching relation between the terminal serial numbers and the wiring content; verifying and optimizing the identified terminal strip names by combining the position areas corresponding to the terminal ranking names to obtain optimized target terminal strip names; integrating the target terminal strip name and the corresponding matching relation between the terminal serial number and the wiring content, and outputting structured terminal strip information; And comparing the structured terminal strip information with preset terminal strip design data, and identifying to obtain difference information.
2. 2. The method of claim 1, wherein the text recognition of the terminal strip image to obtain terminal strip text information comprises: performing text recognition on the terminal strip image by adopting a PaddleOCR model to obtain terminal strip text information, wherein the PaddleOCR model comprises a text detection sub-model and a text recognition sub-model; Correspondingly, the text recognition is performed on the terminal strip image by adopting PaddleOCR model to obtain terminal strip text information, which comprises the following steps: Detecting the terminal strip image by adopting the text detection sub-model so as to locate a text region of an image in the terminal strip image; and identifying the text region by adopting the text identification sub-model to obtain terminal strip text information.
3. 3. The method of claim 1, wherein the acquiring the terminal strip image to be processed comprises: the terminal strip image to be processed is shot and obtained in real time through image acquisition equipment, or the stored terminal strip image is read, wherein the format of the terminal strip image comprises at least one of JPG, PNG, BMP; and performing at least one of graying, noise reduction, image enhancement and geometric correction on the acquired terminal strip image.
4. 4. The method according to claim 1, wherein establishing a correspondence between terminal numbers and wiring contents based on the position coordinates includes: Calculating the space distance between the position coordinates corresponding to the terminal serial numbers and the position coordinates corresponding to the wiring contents; and establishing a corresponding matching relation between the terminal serial numbers of which the space distances are smaller than a preset distance threshold value and the wiring content.
5. 5. The method of claim 1, wherein the location area corresponding to the terminal strip name includes an area in which the text of the terminal strip name identified by PaddleOCR is located, and a terminal ranking area detected by YOLOv 8; And verifying and optimizing the identified terminal strip names by combining the corresponding position areas of the terminal ranking names to obtain optimized target terminal strip names, wherein the method comprises the following steps: Determining the overlapping degree of the area where the terminal ranking text is located and the terminal ranking area; And if the overlapping degree is lower than a preset threshold value, invoking PaddleOCR to perform text recognition on the terminal strip name area detected by YOLOv to obtain an optimized target terminal strip name.
6. 6. And if the overlapping degree is greater than or equal to a preset overlapping degree threshold value, taking the terminal strip name as the target terminal strip name.
7. 7. The method of claim 1, wherein integrating the target terminal strip name and the corresponding matching relationship between the terminal serial number and the wiring content, and outputting structured terminal strip information, comprises: And integrating the target terminal strip name and the corresponding matching relation between the terminal serial number and the wiring content according to a preset data format, and outputting structured terminal strip information, wherein the preset data format comprises at least one of JSON and XML.
8. 8. The method of claim 1, wherein comparing the structured strip information with preset strip design data, and identifying difference information comprises: Comparing whether the structured terminal strip information is consistent with the terminal serial number under the same terminal strip name and the corresponding wiring content in the preset terminal strip design data to obtain difference information, The difference information comprises unmatched terminal serial numbers, wiring content, missing terminal serial numbers or redundant terminal serial numbers.
9. 9. An intelligent recognition device of terminal strip information, which is characterized by comprising: The first acquisition module is used for acquiring a terminal strip image to be processed; The identification module is used for carrying out text identification on the terminal strip image to obtain terminal strip text information, and the terminal strip text information at least comprises a terminal strip name, a terminal sequence number and wiring content; The segmentation module is used for carrying out target mask segmentation on the terminal strip image and outputting position coordinates corresponding to the terminal strip text information, wherein the position coordinates are represented by vertex coordinates of a rectangular frame and at least comprise coordinate information of the upper left corner and the lower right corner of the rectangular frame; The matching module is used for establishing a corresponding matching relation between the terminal serial number and the wiring content based on the position coordinates; The optimizing module is used for verifying and optimizing the identified terminal strip names by combining the position areas corresponding to the terminal ranking names to obtain optimized target terminal strip names; The integrating module is used for integrating the corresponding matching relation between the target terminal strip name and the terminal serial number and the wiring content and outputting structured terminal strip information; And the comparison module is used for comparing the structured terminal strip information with preset terminal strip design data and identifying to obtain difference information.
10. 10. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the steps of the method according to any one of claims 1 to 7 when the computer program is executed.
11. 11. A computer readable storage medium for storing a computer program, the computer program being loaded by a processor to perform the steps of the method of any one of claims 1 to 7.

Description

Intelligent identification method, device, equipment and storage medium for terminal strip information Technical Field The present application relates to the field of image processing technologies, and in particular, to an intelligent terminal strip information identification method, an intelligent terminal strip information identification device, an electronic device, and a computer readable storage medium. Background The terminal strip is a key component for connecting electrical equipment in the power system, and the surface of the terminal strip is marked with information such as a terminal strip name, a terminal serial number, wiring content and the like, and the information is required to be strictly consistent with terminal strip design data in the SDD, so that the installation, the debugging and the maintenance of subsequent equipment are facilitated. The acquisition and verification of the traditional terminal strip information generally depend on manual recording and manual comparison, and the traditional terminal strip information is low in efficiency and is easily affected by human errors. In the prior art, part of schemes automatically recognize text information of terminal blocks by adopting an Optical Character Recognition (OCR) technology, but the defects that the Chinese characters in the terminal block images are densely arranged, the background is complex (such as dirt and light reflection exist), the OCR is easy to recognize errors, and the correlation degree between the terminal serial numbers and the space positions of wiring contents is high. In addition, the traditional OCR can only identify text contents, and cannot realize automatic binding of the text contents and the text contents, and manual intervention is needed. Disclosure of Invention The application provides an intelligent recognition method, an intelligent recognition device, electronic equipment and a computer readable storage medium for terminal strip information, which can realize full-flow automation from recognition to verification of the terminal strip information through text recognition, position positioning and automatic comparison with design data. In order to achieve the above purpose, the application adopts the following technical scheme: in a first aspect, an intelligent recognition method for terminal strip information is provided, including: acquiring a terminal strip image to be processed; text recognition is carried out on the terminal strip image to obtain terminal strip text information, wherein the terminal strip text information at least comprises a terminal strip name, a terminal sequence number and wiring content; performing target mask segmentation on the terminal strip image, and outputting position coordinates corresponding to the terminal strip text information, wherein the position coordinates are represented by vertex coordinates of a rectangular frame and at least comprise coordinate information of the upper left corner and the lower right corner of the rectangular frame; based on the position coordinates, establishing a corresponding matching relation between the terminal serial numbers and the wiring content; verifying and optimizing the identified terminal strip names by combining the position areas corresponding to the terminal ranking names to obtain optimized target terminal strip names; integrating the target terminal strip name and the corresponding matching relation between the terminal serial number and the wiring content, and outputting structured terminal strip information; And comparing the structured terminal strip information with preset terminal strip design data, and identifying to obtain difference information. In a second aspect, an intelligent recognition device for terminal strip information is provided, including: The first acquisition module is used for acquiring a terminal strip image to be processed; The identification module is used for carrying out text identification on the terminal strip image to obtain terminal strip text information, and the terminal strip text information at least comprises a terminal strip name, a terminal sequence number and wiring content; The segmentation module is used for carrying out target mask segmentation on the terminal strip image and outputting position coordinates corresponding to the terminal strip text information, wherein the position coordinates are represented by vertex coordinates of a rectangular frame and at least comprise coordinate information of the upper left corner and the lower right corner of the rectangular frame; The matching module is used for establishing a corresponding matching relation between the terminal serial number and the wiring content based on the position coordinates; The optimizing module is used for verifying and optimizing the identified terminal strip names by combining the position areas corresponding to the terminal ranking names to obtain optimized target terminal strip names; The integrating m