Method and system for judging production progress of shipbuilding intermediate product

Abstract

The invention provides a method and a system for judging the production progress of a shipbuilding intermediate product. The method for judging the production progress of the shipbuilding intermediate product comprises the following steps: collecting a plurality of production site images; the plurality of production site images comprise intermediate products in different process link types; marking the plurality of production site images, and constructing and training an intermediate product production progress recognition model based on the marked production site images; marking the plurality of production site images, wherein the product types of intermediate products of the production site images, the positions of the intermediate products and the types of the process links of the intermediate products are marked; obtaining a to-be-recognized production site image; and inputting the to-be-identified production site image into the trained intermediate product production progress identification model, wherein the trained intermediate product production progress identification model can output the product type of the intermediate product of the to-be-identified production site image, the position of the intermediate product and the process link type of the intermediate product.

Description

Technical Field

[0001] This application belongs to the technical field of shipbuilding, and particularly relates to a method and system for judging the production progress of intermediate products in shipbuilding. Background Art

[0002] Intermediate products in shipbuilding (hereinafter referred to as intermediate products) are stage-by-stage construction products formed by installing multiple outfitting parts on the basis of the hull structure. Outfitting parts include pipes, brackets, electrical outfitting parts, air ducts, etc. The update of the production progress of traditional intermediate products relies on the method of manual work reporting. Work reporting is carried out by the production department during the production execution process to track the production progress. However, there are the following problems with manual work reporting: the progress information is reported late or omitted, data entry may be incorrect, and there is a situation of deliberately hiding the true progress. These problems will affect the accuracy and timeliness of production progress data. Summary of the Invention

[0003] One of the purposes of this application is to provide a method and system for judging the production progress of intermediate products in shipbuilding, which can improve the accuracy of production progress data and ensure the timeliness of production progress data.

[0004] To achieve the above object and other related objects, the first aspect of this application provides a method for judging the production progress of intermediate products in shipbuilding, including the following steps:

[0005] Collect images of multiple production sites; the multiple production site images include intermediate products in different process link types;

[0006] Label the multiple production site images, and construct and train an intermediate product production progress recognition model based on the labeled production site images; labeling the multiple production site images includes labeling the product type of the intermediate product, the position of the intermediate product, and the process link type where the intermediate product is located in the production site image;

[0007] Obtain a production site image to be recognized; input the production site image to be recognized into the trained intermediate product production progress recognition model, and the trained intermediate product production progress recognition model can output the product type of the intermediate product, the position of the intermediate product, and the process link type where the intermediate product is located in the production site image to be recognized.

[0008] The second aspect of this application provides a system for judging the production progress of intermediate products in shipbuilding, including:

[0009] A collection unit for collecting images of multiple production sites; the multiple production site images include intermediate products in different process link types;

[0010] A training unit for annotating the multiple production site images, constructing and training an intermediate product production progress recognition model based on the annotated production site images; annotating the multiple production site images includes annotating the product type of the intermediate product, the position of the intermediate product, and the process link type where the intermediate product is located in the production site image.

[0011] A judgment unit for obtaining a production site image to be recognized; inputting the production site image to be recognized into the trained intermediate product production progress recognition model, and the trained intermediate product production progress recognition model can output the product type of the intermediate product, the position of the intermediate product, and the process link type where the intermediate product is located in the production site image to be recognized.

[0012] This application has at least the following beneficial effects:

[0013] By using the method and system for judging the production progress of intermediate products in shipbuilding of this application, the accuracy of production progress data can be improved and the timeliness of production progress data can be ensured. Description of the Drawings

[0014] In order to more clearly illustrate the technical solutions of the embodiments of this application, the drawings required to be used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application, and therefore should not be regarded as limiting the scope. For those of ordinary skill in the art, without creative efforts, other related drawings can also be obtained based on these drawings.

[0015] Figure 1 It is a schematic flowchart of the method for judging the production progress of intermediate products in shipbuilding shown according to the embodiments of this application. Detailed Embodiments

[0016] The following specific embodiments illustrate the implementation manners of this application. Those skilled in the art can easily understand other advantages and effects of this application from the content disclosed in this specification. This application can also be implemented or applied through other different specific implementation manners. The details of this application can also be variously modified or changed based on different viewpoints and applications without departing from the spirit of this application. It should be noted that, without conflict, the following embodiments and the features in the embodiments can be combined with each other.

[0017] Refer to Figure 1 , the first aspect of this application provides a method for judging the production progress of intermediate products in shipbuilding, including the following steps:

[0018] Collect multiple production site images; the multiple production site images include intermediate products in different process link types.

[0019] The process link types of intermediate products include: large block erection, large block outfitting, pre-outfitting, painting, sub-assembly, sub-assembly outfitting, and sub-assembly painting.

[0020] Different production sites are used to build different intermediate products. The production sites include the internal workshop site and the outdoor open-air site. Correspondingly, collecting production site images includes using surveillance cameras installed on workshop columns, walls, etc. to collect images of the internal workshop production site, and using surveillance cameras installed on outdoor lamp posts or drones to collect images of the outdoor open-air production site.

[0021] In some embodiments, for the situation where the production site area is too large and a single on-line monitoring device cannot cover the entire production site by collecting one image, the production site images can be obtained by two methods: single-device multi-angle image acquisition and stitching, and multi-device multi-angle image acquisition and stitching.

[0022] Single-device multi-angle image acquisition and stitching means using the same on-line monitoring device to take multi-angle photos, and then stitching the images taken at different angles together to form a complete production site image. To ensure consistent stitching effects of multi-angle image acquisition, within a continuous time period, the single device completes the acquisition of all images in the preset angle sequence, which can effectively avoid environmental changes (such as changes in weather conditions or differences in light intensity) caused by shooting at different times, thus preventing problems such as inconsistent brightness or contrast in the final stitched image.

[0023] Multi-device multi-angle image acquisition and stitching means installing multiple on-line monitoring devices around the same production site. These on-line monitoring devices can cover the entire production site. By stitching the images taken by all devices, a complete production site image is finally formed. Multi-device multi-angle image acquisition and stitching preferably uses simultaneous acquisition at the same moment, which can effectively avoid environmental changes (such as changes in weather conditions or differences in light intensity) caused by shooting at different times, thus preventing problems such as inconsistent brightness or contrast in the final stitched image.

[0024] Label multiple production site images, and build and train an intermediate product production progress recognition model based on the labeled production site images; labeling multiple production site images includes labeling the product type of the intermediate product, the position of the intermediate product, and the process link type of the intermediate product in the production site image.

[0025] In some embodiments, the product types of intermediate products include large block erection, section, and sub-assembly.

[0026] In some embodiments, in order to improve the accuracy of labeling the product type of intermediate products and the location of intermediate products, before labeling multiple production site images, it includes image preprocessing of the collected production site images. The image preprocessing includes adjusting the contrast and brightness of the production site images.

[0027] Furthermore, for the production site images collected by the two methods of single-device multi-angle image acquisition and stitching and multi-device multi-angle image acquisition and stitching, the image preprocessing includes de-distortion processing of the stitched production site images to make the shapes of objects in the images more accurate.

[0028] In some embodiments, in order to improve the accuracy of labeling the process step type of intermediate products, before labeling multiple production site images, it includes classifying the collected production site images according to the product type of intermediate products and the process step type of intermediate products.

[0029] Constructing and training an intermediate product production progress recognition model based on the labeled production site images includes: constructing an intermediate product production progress recognition model, using the labeled production site images as a training set, and inputting them into the intermediate product production progress recognition model for training to obtain a trained intermediate product production progress recognition model.

[0030] Obtain a production site image to be recognized; input the production site image to be recognized into the trained intermediate product production progress recognition model, and the trained intermediate product production progress recognition model can output the product type of the intermediate product, the location of the intermediate product, and the process step type of the intermediate product in the production site image to be recognized.

[0031] In some embodiments, the production site image to be recognized can be a stitched production site image. The stitched production site image can be collected by using the above-mentioned single-device multi-angle image acquisition and stitching method or multi-device multi-angle image acquisition and stitching method.

[0032] In some embodiments, the production site image to be recognized can be collected at regular intervals to avoid the situation of late reporting or missed reporting of the production progress of intermediate products mentioned in the background art.

[0033] A preset production management system is used to store the product type, number, location of intermediate products, and the process step type of intermediate products. The product type of intermediate products, the location of intermediate products, and the process step type of intermediate products are stored in the production scheduling data of the production management system.

[0034] To determine that the intermediate products in the production site images output by the intermediate product production progress recognition model are consistent with the product types, positions of the intermediate products, and the types of process links where the intermediate products are located in the preset production management system, and to avoid errors in the input data mentioned in the background art, the method for judging the production progress of intermediate products in shipbuilding of the present application further includes:

[0035] Input the product type and position of the intermediate products in the production site images output by the intermediate product production progress recognition model into the preset production management system, and the production management system can output the numbers of the intermediate products; compare the type of process link of the intermediate products output by the production management system with the type of process link of the intermediate products output by the intermediate product production progress recognition model. If the types of process links are consistent, it indicates that the product type, position of the intermediate products, and the type of process link where the intermediate products are located in the production site images output by the intermediate product production progress recognition model are correct. If the types of process links are inconsistent, update the type of process link of the intermediate products with the corresponding numbers in the production management system. If there is no intermediate product at the corresponding position in the production management system, create a new product type of the intermediate products in the production management system, and store the numbers, positions, and the types of process links where the intermediate products corresponding to the names of the newly created intermediate products are located.

[0036] Furthermore, to avoid wrongly creating the product types of the intermediate products in the production site images in the production management system, the method for judging the production progress of intermediate products in shipbuilding of the present application further includes: performing manual verification before creating the new product type of the intermediate products.

[0037] Furthermore, to improve the accuracy of the intermediate product production progress recognition model, referring to Figure 1 the method for judging the production progress of intermediate products in shipbuilding of the present application further includes updating the intermediate product production progress recognition model. Specifically, it includes marking the product type, position of the intermediate products, and the type of process link where the intermediate products are located on the corresponding production site images, and inputting the marked production site images into the intermediate product production progress recognition model for training.

[0038] The second aspect of the present application provides a system for judging the production progress of intermediate products in shipbuilding, including:

[0039] An acquisition unit for acquiring a plurality of production site images; the plurality of production site images include intermediate products in different types of process links;

[0040] A training unit for annotating the multiple production site images, constructing and training an intermediate product production progress recognition model based on the annotated production site images; annotating the multiple production site images includes annotating the product type of the intermediate product, the position of the intermediate product, and the type of process link where the intermediate product is located in the production site images.

[0041] A judgment unit for obtaining a production site image to be recognized; inputting the production site image to be recognized into the trained intermediate product production progress recognition model, and the trained intermediate product production progress recognition model can output the product type of the intermediate product, the position of the intermediate product, and the type of process link where the intermediate product is located in the production site image to be recognized.

[0042] The above are only the preferred embodiments of the present application. It should be noted that for those of ordinary skill in the art, without departing from the technical principle of the present application, several improvements and substitutions can be made, and these improvements and substitutions should also be regarded as the protection scope of the present application.

Claims

1. A method for determining the production progress of intermediate products in shipbuilding, characterized in that: The steps include: Collecting multiple production site images; the multiple production site images include intermediate products in different process link types; Annotating the multiple production site images, and constructing and training an intermediate product production progress recognition model based on the annotated production site images; annotating the multiple production site images includes annotating the product type of the intermediate products in the production site images, the location of the intermediate products, and the type of process link in which the intermediate products are located; Acquire the image of the production site to be identified; input the image of the production site to be identified into the trained intermediate product production progress identification model, and the trained intermediate product production progress identification model can output the product type of the intermediate product in the image of the production site to be identified, the location of the intermediate product, and the type of process link in which the intermediate product is located.

2. The method for determining the production progress of intermediate products in shipbuilding according to claim 1, characterized in that: The process link types include large assembly, large assembly outfitting, pre-outfitting, painting, general assembly, general assembly outfitting, and general assembly painting.

3. The method for determining the production progress of intermediate products in shipbuilding according to claim 1, characterized in that: The product types of the intermediate products include large assemblies, segments, and complete segments.

4. The method for determining the production progress of intermediate products in shipbuilding according to claim 1, characterized in that: The production site includes an internal site of a workshop and an outdoor site. The collecting of images of the production site includes using a surveillance camera to collect images of the internal production site of the workshop, and using a surveillance camera or a drone to collect images of the outdoor production site.

5. The method for determining the production progress of intermediate products in shipbuilding according to claim 1, characterized in that: The method of collecting production site images includes at least one of a single-device multi-angle image collection and stitching method and a multi-device multi-angle image collection and stitching method.

6. The method for determining the production progress of intermediate products in shipbuilding according to claim 1, characterized in that: Before labeling the multiple production site images, the collected production site images are preprocessed, and the image preprocessing includes adjusting the contrast and brightness of the production site images.

7. The method for determining the production progress of intermediate products in shipbuilding according to claim 1, characterized in that: Before labeling the multiple production site images, the collected production site images are classified according to the product types of the intermediate products and the types of process links in which the intermediate products are located.

8. The method for determining the production progress of intermediate products in shipbuilding according to claim 1, characterized in that: The method of constructing and training an intermediate product production progress recognition model based on the annotated production site images includes: constructing an intermediate product production progress recognition model, inputting the annotated production site images as training sets into the intermediate product production progress recognition model for training, and obtaining a trained intermediate product production progress recognition model.

9. The method for determining the production progress of intermediate products in shipbuilding according to claim 1, characterized in that: The preset production management system is used to store the product type, number, location of the intermediate product and the type of process link of the intermediate product. The shipbuilding intermediate product production progress judgment method also includes: inputting the product type and location of the intermediate product of the production site image output by the intermediate product production progress recognition model into the preset production management system, and the production management system can output the number of the intermediate product; comparing the process link type of the intermediate product output by the production management system with the process link type of the intermediate product output by the intermediate product production progress recognition model: If they are consistent, it means that the product type, location and process link type of the intermediate product in the production site image output by the intermediate product production schedule recognition model are correct; If the process link types are inconsistent, updating the process link types of the intermediate products with corresponding numbers in the production management system; If there is no intermediate product at the corresponding position in the production management system, a new product type of the corresponding intermediate product is created in the production management system, and the number, position and process link type of the intermediate product corresponding to the newly created product type of the intermediate product are stored.

10. The method for determining the production progress of intermediate products in shipbuilding according to claim 9, characterized in that: The product type, location and process link type of the newly created intermediate product are marked in the corresponding production site image, and the marked production site image is input into the intermediate product production progress recognition model for training.

11. A system for determining the production progress of intermediate products in shipbuilding, characterized in that: include: A collection unit, used for collecting images of multiple production sites; The multiple production site images include intermediate products in different process link types; A training unit is used to annotate the multiple production site images, and to construct and train an intermediate product production progress recognition model based on the annotated production site images; the annotating of the multiple production site images includes annotating the product type of the intermediate product in the production site images, the location of the intermediate product, and the type of process link in which the intermediate product is located; The judgment unit is used to obtain the image of the production site to be identified; the image of the production site to be identified is input into the trained intermediate product production progress recognition model, and the trained intermediate product production progress recognition model can output the product type of the intermediate product in the image of the production site to be identified, the location of the intermediate product and the type of process link in which the intermediate product is located.

Images