A sub-pixel detection method for structural cracks

Abstract

The invention relates to a sub-pixel detection method for cracks in a structure, comprising the following steps: 1. Obtain multiple surface images of the structure, and record the position and attitude information of the camera when each image is taken; 2. According to the lens parameters of the camera when shooting Image correction and image calibration are performed on the captured images with the position and attitude information to obtain the calibration coefficient of the distance from the camera to the shooting plane; 3. Stitching multiple images taken to obtain the surface expansion diagram of the structure; 4. Based on training A good crack image training model, which divides the surface expansion diagram of the structure into regions, and extracts cracks in different regions; 5. Thinning the extracted crack binary image, obtaining the crack skeleton map, and calculating the starting point position of the crack Information, end point position information, bifurcation point position information, and calculate the sub-pixel width and length of the crack. The invention takes both efficiency and precision into account, and can detect crack parameters of 0.1 pixel.

Description

technical field [0001] The present invention relates to the technical field of crack sub-pixel detection in structures, in particular to a sub-pixel detection method for cracks in structures. Background technique [0002] Any architectural structure will be affected by the natural environment and human activities, resulting in varying degrees of degradation and decay. Over time, these visible or subtle changes may affect the safety and durability of the structure. In some special fields, such as water conservancy projects, cracks can endanger the safety of dams; in nuclear power plants, defects in the concrete casing of reactors can cause radiation leakage, and defects in air cooling towers can affect production. For some large modern architectural structures such as stadiums, terminal buildings and grand theaters, the decline of the structure is directly related to personal safety. [0003] At present, in China, external methods are mainly used for surface inspection of st...

AI Technical Summary

Problems solved by technology

The traditional edge detection algorithm can only realize the whole pixel detection, that is, it can only detect the edge of the unit pixel level.

The traditional method can only identify crack images whose crack width is equal to or greater than 2 pixels. If the crack width is less than 2 pixels, this method cannot identify cracks, nor can it realize the recognition of sub-pixel crack images.

Images