A method for identifying sub-images in a plan or section staircase detail drawing of an architectural specialty

Abstract

The present invention discloses a method for identifying sub-images in a plane or section staircase drawing of a building specialty, and the method comprises the following steps: S1, drawing analysis to obtain the basic information of the graphic element and the layer; S2, the basic information obtained, the wall line layer; S3, image binarization; S4, image processing, finding all contours; S5, contour screening, eliminating the smaller ones; S6, judging the selected contours, if only one space contour is obtained and the area of ​​this contour is greater than 70% of the area of ​​the frame, it means that this contour is a large frame and not a sub-image; S7, removing the remaining image of the large frame, performing another contour search, and finally obtaining the sub-image contour list, and saving it in a dictionary. The present invention can accurately and stably obtain all the sub-image information in the plane or section staircase drawing, thereby providing good conditions for the rule review related to the sub-image in the plane or section staircase drawing.

Description

Technical Field

[0001] The present invention relates to the fields of CAD drawing recognition, computer vision technology, and image processing technology, and in particular to a method for recognizing sub-pictures in a plane or section staircase drawing of a building specialty. Background Art

[0002] CAD construction drawings are drawings made by using design software such as AutoCAD to show the overall layout of the project, the exterior shape of the building, the interior layout, the structural structure, the interior and exterior decoration, the material methods, the equipment, and the construction. CAD construction drawings are characterized by complete drawings, accurate expressions, and specific requirements. They are the basis for engineering construction, the preparation of construction drawing budgets, and the construction organization design. They are also important technical documents for technical management. Before construction, the construction drawings need to be carefully reviewed before entering the construction stage. The purpose is to ensure the smooth progress of the construction and avoid the impact of errors in the drawings on the use stage after the completion of the construction.

[0003] The architectural drawing in the CAD construction drawing is a drawing that shows the shape, size, material and method of the details or building components and accessories of the building in a larger scale (generally 1:20, 1:10, 1:5, etc.). Specifically, the details of the local structure of the house that need to be clearly reflected are drawn in a larger scale to express the construction method, size, relationship between components and accessories, and building materials. The architectural drawing can be used during the construction process, not for the overall design. The architectural drawing can be scaled arbitrarily and will not be blurred. If the drawings in ordinary format are enlarged too much, they will appear unclear. The reason for the need to enlarge is to see the details clearly during the construction process, rather than just understanding the rough information such as the layout from the drawing.

[0004] With the widespread application of artificial intelligence, some tasks that were previously done manually can be completed with artificial intelligence. The review of CAD construction drawings is a time-consuming and labor-intensive repetitive task, and it is easy for reviewers to miss them. Artificial intelligence can identify sub-images in CAD construction drawings. With the help of computer vision technology and traditional image processing algorithms, accurate sub-image recognition can be achieved, thereby realizing automatic review of sub-image related information, such as whether the handrails next to the stairs are installed in a reasonable position. The present invention proposes a recognition method with good generalization performance and high accuracy for obtaining sub-image information in architectural professional plan / section staircase drawings. Summary of the invention

[0005] In view of the above problems, the present invention provides a method for identifying sub-images in a plan or section staircase drawing of a building specialty.

[0006] To solve the above technical problems, the present invention is implemented by the following technical solutions: The present invention provides a method for identifying sub-images in a plan or section staircase drawing of a building specialty, the method comprising the following steps:

[0007] S1. First, the CAD drawing needs to be parsed to obtain the basic information of the elements and layers in the drawing;

[0008] S2, then obtaining the wall line layer according to the basic information obtained in step S1;

[0009] S3, binarizing the image obtained in step S2;

[0010] S4, performing image processing on the binary image in step S3 to find all the contours, i.e., closed areas composed of pixels;

[0011] S5, screening the group of contours obtained in step S4, removing those with a value less than a specific threshold, and retaining only a few contours with a value greater than the specific threshold;

[0012] S6, judging the outline selected in step S5, if only one spatial outline is obtained and the area of ​​this outline is greater than 70% of the area of ​​the frame, it means that this outline is a large frame and not a sub-image;

[0013] S7, performing another contour search on the image remaining after removing the large frame in step S6 to finally obtain a list of sub-image contours, and convert the contours into outer frames and save them in a dictionary.

[0014] Preferably, after the wall line layer is acquired in step S2, the image information needs to be retained.

[0015] Preferably, after binarizing the image in step S3, only pixels within a specific range are retained.

[0016] Preferably, after judging step S5, step S6 further comprises: if there is more than one contour, it means that the contour is a sub-image, and the contour is converted into an outer frame and stored in the dictionary.

[0017] Preferably, the result of the sub-image recognition in step S7 is stored in a dictionary for use in reviewing the image.

[0018] It can be seen from the above description of the present invention that, compared with the prior art, the present invention has the following advantages:

[0019] The present invention provides a method for identifying sub-images in a plan or section staircase drawing of a building specialty, which can accurately and stably obtain all sub-image information in the plan or section staircase drawing, thereby providing good conditions for reviewing rules related to sub-images in the plan or section staircase drawing. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 It is a flow chart of the steps of the present invention;

[0021] Figure 2 is a flowchart of specific steps of an embodiment of the present invention;

[0022] Figure 3 It is a schematic diagram of an embodiment of the present invention accurately identifying a sub-image on a CAD drawing. DETAILED DESCRIPTION

[0023] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0024] like Figure 1-2 As shown, the present invention provides a method for identifying sub-images in a plan or section staircase drawing of a building specialty, the method comprising the following steps:

[0025] S1. First, the CAD drawing needs to be parsed to obtain the basic information of the elements and layers in the drawing;

[0026] S2, then obtaining the wall line layer according to the basic information obtained in step S1;

[0027] S3, binarizing the image obtained in step S2;

[0028] S4, performing image processing on the binary image in step S3 to find all the contours, i.e., closed areas composed of pixels;

[0029] S5, screening the group of contours obtained in step S4, removing those with a value less than a specific threshold, and retaining only a few contours with a value greater than the specific threshold;

[0030] S6, judging the outline selected in step S5, if only one spatial outline is obtained and the area of ​​this outline is greater than 70% of the area of ​​the frame, it means that this outline is a large frame and not a sub-image;

[0031] S7, performing another contour search on the image remaining after removing the large frame in step S6 to finally obtain a list of sub-image contours, and convert the contours into outer frames and save them in a dictionary.

[0032] Among them, after the wall line layer is obtained in step S2, the image information needs to be retained.

[0033] Among them, after the image is binarized in step S3, only pixels within a specific range are retained.

[0034] Among them, after judging step S5, step S6 also includes: if there is more than one contour, it means that the contour is a sub-image, and the contour is converted into an outer frame and stored in the dictionary.

[0035] The result of the sub-image recognition in step S7 is stored in the dictionary for use in image review.

[0036] like Figure 1-2 As shown, the present invention provides a method for identifying sub-images in a plan or section staircase drawing of a building specialty, and the specific method includes the following steps:

[0037] 1. Find the contour by adjusting the gray value of the image and retain the contours with an area greater than 100,000;

[0038] 2. Judge the retained contours. If there is only one contour and the ratio of this contour to the frame is greater than 0.7, then this contour is not a sub-image area. If not, the sub-image search is completed and saved in the dictionary;

[0039] 3. Delete the frame for non-sub-image contours and re-execute the first step until a suitable sub-image frame is found and saved in the dictionary.

[0040] The present invention provides a method for identifying sub-pictures in a plan or section staircase drawing of a building specialty. The method can identify the information of sub-pictures in a drawing when the sub-pictures exist in various forms. The method has good generalization and high accuracy. The identified sub-picture information is convenient for checking it according to building specifications during drawing review, such as checking whether the handrails next to the stairs are installed in a reasonable position.

[0041] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for identifying sub-images in a plan or section staircase drawing of a building, characterized in that: The method comprises the following steps: S1. First, the CAD drawing needs to be parsed to obtain the basic information of the elements and layers in the drawing; S2, then obtain the wall line layer according to the basic information obtained in step S1; S3, binarizing the image obtained in step S2; S4, performing image processing on the binary image in step S3 to find all the contours, i.e., closed areas composed of pixels; S5, screening the group of contours obtained in step S4, removing those with a value less than a specific threshold, and retaining only a few contours with a value greater than the specific threshold; S6, judging the outline selected in step S5, if only one spatial outline is obtained and the area of ​​this outline is greater than 70% of the area of ​​the frame, it means that this outline is a large frame and not a sub-image; S7, performing another contour search on the image remaining after removing the large frame in step S6 to finally obtain a list of sub-image contours, and convert the contours into outer frames and save them in a dictionary.

2. According to claim 1, a method for identifying sub-images in a plan or section staircase drawing of a building, characterized in that: After the wall line layer is obtained in step S2, the image information needs to be retained.

3. According to claim 1, a method for identifying sub-images in a plan or section staircase drawing of a building, characterized in that: After the image is binarized in step S3, only pixels within a specific range are retained.

4. According to claim 1, a method for identifying sub-images in a plan or section staircase drawing of a building, characterized in that: After the step S5 is judged in the step S6, the step further includes: if there is more than one contour, the contour is converted into an outer frame and stored in the dictionary.

5. According to claim 1, a method for identifying sub-images in a plan or section staircase drawing of a building, characterized in that: The result of the sub-image recognition in step S7 is stored in the dictionary for use in image review.

Images