CN-117036298-B - Crack information identification method and computer equipment

Abstract

The disclosure relates to a crack information identification method and computer equipment. The method comprises the steps of obtaining an image to be detected, carrying out crack identification and image segmentation on the image to be detected, determining a crack identification result in the image to be detected, determining length information of a crack in the image to be detected based on the crack identification result and a skeleton method, determining width information of the crack in the image to be detected based on the crack identification result and a structural element corrosion method, and determining the information of the crack in the image to be detected according to the crack identification result, the length information of the crack and the width information of the crack. By adopting the method, the morphological characteristics of the crack can be accurately identified, and the length and the width of the crack can be accurately detected.

Inventors

• LIU YUKAI
• CAO YAN
• WANG HONGFEI

Assignees

• 苏州光格科技股份有限公司

Dates

Publication Date

20260508

Application Date

20230815

Claims (8)

1. 1. A method of identifying crack information, the method comprising: Acquiring an image to be detected; performing crack identification and image segmentation on the image to be detected, and determining a crack identification result in the image to be detected; Performing pixel assignment on the crack identification result to obtain a binarized image, wherein the pixel assignment of a crack position in the binarized image is 1, and the pixel assignment of a non-crack position is 0; determining the length information of the crack in the image to be detected based on the crack identification result and a skeleton method; corroding crack positions in the binary image by using a predetermined initial structural element; In response to the fact that reserved pixel points do not exist in the corroded crack position, reducing the size of the initial structural element until reserved pixel points appear in the corroded crack position after the crack position is corroded by the reduced initial structural element; determining width information of a crack in the image to be detected based on the reduced initial structural element corresponding to the reserved pixel point, wherein the initial structural element is circular in shape; and determining the information of the crack in the image to be detected according to the crack identification result, the length information of the crack and the width information of the crack.
2. 2. The method according to claim 1, wherein determining the length information of the crack in the image to be detected based on the crack identification result and a skeleton method includes: determining a crack image to be thinned according to the assignment result of the pixels in the binarized image; for each pixel point of the crack image to be thinned, determining eight neighborhood points of each pixel point; Determining pixel points which do not meet preset conditions according to each pixel point and eight neighborhood points of each pixel point to obtain target pixel points, wherein the preset conditions are determined based on the eight neighborhood points; And determining the length information of the crack in the image to be detected based on the target pixel point.
3. 3. The method according to claim 1, wherein the performing crack recognition and image segmentation on the image to be detected, and determining the crack recognition result in the image to be detected, includes: Detecting a crack region of each crack in the image to be detected, wherein a plurality of marking frames are used in the crack region of each crack to carry out superposition marking from a starting end of the crack region to a terminal end of the crack region; Image segmentation is carried out on each labeling frame, and cracks in each labeling frame are identified; and determining a crack identification result in the image to be detected based on the cracks in each labeling frame.
4. 4. The method according to claim 1, wherein after determining the information of the crack in the image to be detected according to the crack identification result, the length information of the crack, and the width information of the crack, the method further comprises: Acquiring shooting parameters when shooting the image to be detected, wherein the shooting parameters comprise the distance from a lens to a shot plane, the angle of view of the lens in the horizontal direction, the angle of view of the lens in the vertical direction, and the width and the height of the image to be detected; determining the size of each pixel in the horizontal direction in the image to be detected and the size of each pixel in the vertical direction in the image to be detected based on the shooting parameters; Determining the actual size of each pixel in the image to be detected based on the size of each pixel in the horizontal direction in the image to be detected and the size of each pixel in the vertical direction in the image to be detected; and determining the actual size information of the cracks based on the actual size of each pixel and the information of the cracks in the image to be detected.
5. 5. The method according to any one of claims 1 to 4, further comprising: Predicting the trend of the crack in the image to be detected based on a gain piecewise linear regression mode.
6. 6. The method of claim 5, wherein predicting the trend of the crack in the image to be detected based on the gain piecewise linear regression method comprises: Dividing the crack in the image to be detected into a plurality of sections of crack intervals; calculating the crack gain of each section of crack interval based on the preset center of gravity of the crack and the crack end points at two ends of the crack in each section of crack interval; Predicting the trend of the crack in the image to be detected at least according to the target regression parameter of the linear regression function of each section of crack interval and the crack gain of each section of crack interval.
7. 7. The method of claim 6, wherein predicting the trend of the fracture in the image to be detected based at least on the target regression parameter of the linear regression function for each segment of the fracture interval and the fracture gain for each segment of the fracture interval comprises: Acquiring a historical image of the crack; Predicting the trend of the crack in the image to be detected based on a target regression parameter of a linear regression function according to each section of crack interval and the crack gain of each section of crack interval in response to the position of the crack in the historical image being the same as the position of the crack in the image to be detected; Determining a new crack position in which the position of the crack in the historical image is different from the position of the crack in the image to be detected in response to the position of the crack in the historical image being different from the position of the crack in the image to be detected; And determining the trend of the newly-increased crack at the newly-increased crack position, and predicting the trend of the crack in the image to be detected based on the trend of the newly-increased crack and the prediction result of the trend of the crack in the historical image.
8. 8. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor implements the steps of the method of any of claims 1 to 7 when the computer program is executed.

Description

Crack information identification method and computer equipment Technical Field The disclosure relates to the technical field of crack detection, in particular to a crack information identification method and computer equipment. Background With the development of industrial technology, more and more buildings or industrial equipment often have problems of aging or damage of service life and the like along with the time. For example, the condition that various precipitate cracks appear on the wall surface of the tunnel often along with the passage of time, and timely detection of the condition of the tunnel is a key for guaranteeing normal production and living order. With the development of computer vision, the scheme for detecting cracks by utilizing a visual mode is gradually applied to industrial scenes instead of a manual detection method. However, the conventional crack detection method cannot accurately identify the morphological characteristics of the crack, and in addition, cannot accurately detect the length and width of the crack. Disclosure of Invention In view of the above, it is necessary to provide a crack information identification method and a computer device capable of accurately identifying morphological features of a crack and accurately detecting the length and width of the crack. In a first aspect, the present disclosure provides a crack information identification method, the method including: Acquiring an image to be detected; performing crack identification and image segmentation on the image to be detected, and determining a crack identification result in the image to be detected; determining the length information of the crack in the image to be detected based on the crack identification result and a skeleton method; Determining width information of the crack in the image to be detected based on the crack identification result and a structural element corrosion method; and determining the information of the crack in the image to be detected according to the crack identification result, the length information of the crack and the width information of the crack. In one embodiment, the determining the length information of the crack in the image to be detected based on the crack identification result and the skeleton method includes: Performing pixel assignment on the crack identification result to obtain a binarized image, wherein the pixel assignment of a crack position in the binarized image is 1, and the pixel assignment of a non-crack position is 0; determining a crack image to be thinned according to the assignment result of the pixels in the binarized image; for each pixel point of the crack image to be thinned, determining eight neighborhood points of each pixel point; Determining pixel points which do not meet preset conditions according to each pixel point and eight neighborhood points of each pixel point to obtain target pixel points, wherein the preset conditions are determined based on the eight neighborhood points; And determining the length information of the crack in the image to be detected based on the target pixel point. In one embodiment, the determining width information of the crack in the image to be detected based on the crack identification result and the structural element corrosion method includes: corroding crack positions in the binarized image by utilizing a predetermined initial structural element; In response to the fact that reserved pixel points do not exist in the corroded crack position, reducing the size of the initial structural element until reserved pixel points appear in the corroded crack position after the crack position is corroded by the reduced initial structural element; and determining width information of the crack in the image to be detected based on the reduced initial structural element corresponding to the reserved pixel point. In one embodiment, the initial structural element is circular in shape. In one embodiment, the performing crack recognition and image segmentation on the image to be detected, and determining the crack recognition result in the image to be detected includes: Detecting a crack region of each crack in the image to be detected, wherein a plurality of marking frames are used in the crack region of each crack to carry out superposition marking from a starting end of the crack region to a terminal end of the crack region; Image segmentation is carried out on each labeling frame, and cracks in each labeling frame are identified; And determining a crack identification result in the image to be detected based on the cracks in each labeling frame. In one embodiment, after the information of the crack in the image to be detected is determined according to the crack identification result, the length information of the crack and the width information of the crack, the method further includes: Acquiring shooting parameters when shooting the image to be detected, wherein the shooting parameters comprise the distance from a le