Production process picture labeling method

Abstract

The invention belongs to the technical field of production process optimization, and particularly discloses a production process picture labeling method which comprises the following steps: S1, constructing a labeling information database which comprises a data process text database and a process picture database; s2, uploading the process picture; s3, pre-processing the picture, and automatically processing the picture through a process picture database; s4, editing and customizing the picture, and manually adjusting the picture through a labeling system; and S5, outputting and updating a result, and outputting and updating the result to the annotation information database. By integrating advanced computer vision and image processing technologies, automatic and accurate labeling of process pictures is realized, so that the process compiling efficiency and quality are greatly improved.

Description

Technical Field

[0001] The invention belongs to the technical field of production process optimization, and specifically relates to a production process picture marking method. Background Art

[0002] In the process of production process compilation, process pictures, as an intuitive and vivid visual aid, are essential for process understanding and execution. However, traditional process picture annotation methods often rely on image editing software operations, which is not only inefficient and inconsistent in annotation form, but also prone to annotation errors due to human factors, thus affecting the smooth progress of the entire production process. In addition, with the increasing complexity and sophistication of production processes, the accuracy and efficiency requirements for process picture annotation are also getting higher and higher, and traditional annotation methods can no longer meet these needs. Summary of the invention

[0003] In view of the problem that process image annotation is difficult and cumbersome in the prior art, a production process image annotation method is proposed. The present invention provides the following technical solutions: A production process image annotation method comprises the following steps: S1. Constructing a labeling information database, which includes a data process text database and a process image database; S2, upload process pictures; S3, image preprocessing, automatically processing images through the process image database; S4, picture editing and customization, manual adjustment of pictures through the annotation system; S5. Output and update the results: output and update the results to the annotation information database.

[0004] Preferably, between step S4 and step S5, there is also step S4.5.1, user preliminary verification: Preview the image. If it does not meet the requirements, return to step S4 for preliminary adjustment.

[0005] Preferably, after step S4.5.1, there is further provided step S4.5.2, the administrator verifies again: If it meets the requirements, proceed to the next step, otherwise return to step S4 and make adjustments again.

[0006] Preferably, in step S1, a process image preprocessing template library is constructed based on the process text database and the process image database, and the process image preprocessing template library includes multiple standard templates. In step S2, the standard template is matched according to the uploaded process image. In step S3, preprocessing is performed according to the standard template.

[0007] Preferably, the image recognition system in the process picture database decomposes an image feature information library. When matching the standard template according to the uploaded process picture, the image features are automatically identified by the image recognition system, and the standard template is matched according to the image feature information library.

[0008] Preferably, the annotation information database also includes an annotation quick information database, and the annotation quick information includes annotation point location information, annotation style information and annotation text information. In step S2, after uploading the picture, the image features are automatically identified by the image recognition system, and the annotation quick information is automatically matched according to the image features and recommended to the user.

[0009] Preferably, in step S4, when the user annotates, the annotation system synonymously expands the actual annotation information and recommends matching annotation shortcut information in real time.

[0010] Preferably, in step S4, the annotation system automatically compares the actual annotation information with the standard template in the process image preprocessing template library, and provides a real-time reminder of the difference.

[0011] Preferably, the marking information database also includes process compilation information, and the process compilation information includes material information. In step S4, the actual marking information and the material information are compared, and if the material information is inconsistent, a prompt is given.

[0012] Preferably, in step S3, when preprocessing the image, initial information of the uploaded image is first obtained, and the initial information and the standard template include format, brightness, file size and dimension information.

[0013] Compared with the prior art, the present invention has the following beneficial effects.

[0014] 1. Accurate annotation: Allow users to customize annotation parameters and accurately add annotation information based on these parameters to reduce human errors.

[0015] 2. Comprehensive functions: It provides comprehensive image editing functions such as cropping, marking, and adding straight lines to meet various process image annotation needs.

[0016] 3. Easy operation: Through an intuitive interface and simple operation process, users can easily get started and use the device and method of the present invention.

[0017] 4. Strong scalability: The method and device of the present invention can be easily integrated into the existing process compilation system to achieve seamless connection with other systems. At the same time, the accuracy and efficiency of labeling can be improved by continuously upgrading and optimizing the software algorithm. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 is a schematic diagram of embodiment 1 of the present invention; Figure 2 is a schematic diagram of embodiment 2 of the present invention; DETAILED DESCRIPTION In order to enable those skilled in the art to better understand the technical solution of the present invention, the technical solution of the present invention is clearly and completely described below in conjunction with the accompanying drawings of the present invention. The directional words mentioned in the following embodiments, such as "up", "down", "left" and "right", etc., are only reference to the directions of the accompanying drawings. Therefore, the directional words used are used to illustrate rather than limit the invention.

[0019] Example 1 like Figure 1 As shown, a production process image annotation method includes the following steps: S1. Constructing a labeling information database, which includes a data process text database and a process image database; S2, upload process pictures; S3, image preprocessing, automatically processing images through the process image database; S4, picture editing and customization, manual adjustment of pictures through the annotation system; S5. Output and update the results: output and update the results to the annotation information database.

[0020] The annotation information database can be used to efficiently and accurately annotate process images, thereby improving the efficiency and quality of process compilation.

[0021] Example 2 like Figure 2 As shown, a production process image annotation method includes the following steps: S1. Construct a labeling information database, which includes a data process text database and a process image database.

[0022] A process picture preprocessing template library is constructed according to the process text database and the process picture database, and the process picture preprocessing template library includes multiple standard templates.

[0023] S2. Upload process pictures.

[0024] Match the standard template according to the uploaded process image; the matching method is: the image recognition system in the process image database decomposes the image feature information library, and when matching the standard template according to the uploaded process image, the image recognition system automatically identifies the image features and matches the standard template according to the image feature information library. It provides a standardized framework for subsequent image preprocessing and editing, effectively improving processing efficiency and consistency of results.

[0025] The annotation information database also includes an annotation quick information library. The annotation quick information includes annotation point location information, annotation style information and annotation text information. After uploading the picture, the image features are automatically identified through the image recognition system, and the annotation quick information is automatically matched according to the image features and recommended to the user.

[0026] It can significantly reduce the time cost of manual screening, further improve labeling efficiency, and reduce the risk of human error; the image recognition system adopts existing mature computer vision and image processing technology, which will not be described in detail.

[0027] S3, image preprocessing, automatically processing images through the process image database.

[0028] Preprocessing was performed according to the standard template.

[0029] When preprocessing an image, first obtain the initial information of the uploaded image. The initial information and the standard template contain format, brightness, file size and dimension information.

[0030] S4, picture editing and customization, manual adjustment of pictures through the annotation system.

[0031] When users annotate, the annotation system will synonymously expand the actual annotation information and recommend matching annotation shortcut information in real time.

[0032] The annotation system automatically compares the actual annotation information with the standard templates in the process image preprocessing template library, and provides real-time reminders of the differences.

[0033] The marking information database also includes process compilation information, and the process compilation information includes material information. In step S4, the actual marking information and the material information are compared. If the material information is inconsistent, a prompt is given.

[0034] S4.5.1. User preliminary verification: Preview the image. If it does not meet the requirements, return to step S4 for preliminary adjustment.

[0035] Step S4.5.2: The administrator verifies again: S5. Output and update the results: output and update the results to the annotation information database.

[0036] If it meets the requirements, proceed to the next step, otherwise return to step S4 and make adjustments again.

[0037] The final quality and reliability of the annotation results are ensured through the dual guarantee of initial verification by users and re-verification by administrators; any areas that do not meet the requirements can be discovered in a timely manner and returned for adjustment, thus avoiding the transmission and accumulation of erroneous information.

[0038] It can be continuously upgraded and optimized to improve the accuracy and efficiency of annotation. With the continuous development of technology, users can get better annotation experience and effect by updating the software version. The present invention realizes the automatic and accurate annotation of process pictures by integrating advanced computer vision and image processing technology, thereby greatly improving the efficiency and quality of process compilation.

[0039] Example 3 Based on Example 2, the annotation system includes tools such as cropping, marking, and adding straight lines. Users can set the style and layout of annotations by defining image editing parameters. For example, users can choose whether to enable the automatic numbering function, select the direction of the annotation (left or right), set the width of the line, etc. The setting of these parameters can help users control the appearance and readability of the annotation more flexibly.

[0040] During the editing process, users can complete editing tasks such as cropping, adding marks and straight lines by simply dragging and clicking. The system provides an intuitive interface and friendly operation prompts, allowing users to easily get started even if they have no image processing experience.

[0041] After completing the image editing, users can start adding annotation information such as numbers and text descriptions. Users can click once at the starting point of the mark, and then click again at the location where the corner is required. The system will automatically add annotation lines and numbers according to the annotation direction set by the user. If the automatic numbering function is enabled, the system will also automatically generate numbering information according to the preset rules. Users can also manually enter numbers or other text descriptions as needed.

[0042] The system also supports adding multiple sets of annotation information. Users can add multiple annotation points and lines to the image by clicking the "Add" button multiple times. In addition, users can also modify or undo previous editing and annotation operations through functions such as "Undo", "Redo" and "Reset".

[0043] After completing the annotation, users can choose to save or export the annotation results. The system supports saving the annotated images in multiple formats (such as JPEG, PNG, etc.), which is convenient for users to view and use on different platforms and devices. At the same time, the system also provides an export function, and users can export the annotation results to a specific file format (such as base64, SVG, etc.) for further processing and analysis in other software or systems.

[0044] It can be easily integrated into the existing process compilation system and seamlessly connected with other functional modules. By connecting with the interface of the existing system, data sharing and exchange can be realized. Users can share the annotated pictures or data with team members or relevant departments to promote the exchange and collaboration of process information.

Claims

1. A production process picture annotation method, characterized in that: The following steps are involved: S1. Constructing a labeling information database, which includes a data process text database and a process image database; S2, upload process pictures; S3, image preprocessing, automatically processing images through the process image database; S4, picture editing and customization, manual adjustment of pictures through the annotation system; S5. Output and update the results: output and update the results to the annotation information database.

2. The production process image annotation method according to claim 1, characterized in that: Between step S4 and step S5, there is also step S4.5.1, preliminary user verification: Preview the image. If it does not meet the requirements, return to step S4 for preliminary adjustment.

3. The production process image annotation method according to claim 2, characterized in that: After step S4.5.1, there is also step S4.5.2, the administrator verifies again: If it meets the requirements, proceed to the next step, otherwise return to step S4 and make adjustments again.

4. The production process image annotation method according to claim 1, characterized in that: In step S1, a process image preprocessing template library is constructed based on the process text database and the process image database. The process image preprocessing template library includes multiple standard templates. In step S2, the standard template is matched according to the uploaded process image. In step S3, preprocessing is performed according to the standard template.

5. The production process image annotation method according to claim 4, characterized in that: The image recognition system in the process picture database decomposes the image feature information library. When matching the standard template according to the uploaded process picture, the image recognition system automatically identifies the image features and matches the standard template according to the image feature information library.

6. The production process image annotation method according to claim 4 or 5, characterized in that: The annotation information database also includes an annotation quick information database. The annotation quick information includes annotation point location information, annotation style information and annotation text information. In step S2, after uploading the picture, the image features are automatically identified through the image recognition system, and the annotation quick information is automatically matched according to the image features and recommended to the user.

7. The production process image annotation method according to claim 6, characterized in that: In step S4, when the user annotates, the annotation system will synonymously expand the actual annotation information and recommend matching annotation shortcut information in real time.

8. The production process image annotation method according to claim 4, characterized in that: In step S4, the annotation system automatically compares the actual annotation information with the standard template in the process image preprocessing template library, and reminds the difference in real time.

9. The production process image annotation method according to claim 1 or 8, characterized in that: The marking information database also includes process compilation information, and the process compilation information includes material information. In step S4, the actual marking information and the material information are compared. If the material information is inconsistent, a prompt is given.

10. The production process image annotation method according to claim 4, characterized in that: In step S3, during the image preprocessing, the initial information of the uploaded image is first obtained, and the initial information and the standard template include format, brightness, file size and dimension information.

Images