An Intelligent Matching Detection Method for Tunnel Cracks

Abstract

The invention discloses an intelligent matching detection method for tunnel cracks, which comprises the following steps: performing ID numbering on each extracted crack area, then extracting crack feature values ​​in the crack area, establishing a tunnel crack database, and then converting the collected The tunnel crack picture is preprocessed, and then the crack detection is performed on the B picture to obtain the coordinates of the crack area, and then the crack area is quickly positioned and the eigenvalue is extracted through the crack matching network to obtain the eigenvalue vector, and then the eigenvalue vector is compared with the database Compare the crack eigenvalues ​​corresponding to each ID number, confirm the ID number of the output eigenvalue, and finally extract the crack binary image and crack width of the C crack area, and store them in the database together with the ID number to complete the tunnel crack analysis. match detection. The invention has the characteristics of high matching efficiency and good matching accuracy.

Description

technical field [0001] The invention relates to a detection method for tunnel cracks, in particular to an intelligent matching detection method for tunnel cracks. Background technique [0002] Tunnels are an important component of transportation infrastructure, but after they are put into use, a large number of cracks of different shapes will be produced due to human or natural factors. Operators need to identify and detect these cracks to determine whether the cracks will affect the tunnel hole. cause further damage inside the body and threaten driving safety. [0003] However, the current detection methods for cracks have different degrees of defects. For example, in the patent 201610551362.9 "Method for Intelligent Identification and Classification of Cracks", it is proposed to use OpenCV to read the grayscale image and convert it into a histogram, and then segment the crack through the histogram The area is merged with the crack template, and then the crack template is ...

AI Technical Summary

Problems solved by technology

However, this method requires a line array CCD camera to be equipped with a scanning motion when acquiring a two-dimensional image, and a device such as a grating is used to record the coordinates of each scanning line of the CCD camera, and to determine the corresponding position of each pixel of the image on the measured object. position, resulting in long overall image acquisition time and low measurement efficiency; and due to the increased system complexity and cost due to the existence of scanning motion and corresponding position feedback links, the image accuracy may be affected by the scanning motion accuracy. As a result, when the tunnel is uneven or has steps, it is difficult to obtain accurate position information and the measurement effect cannot be achieved

[0006] Therefore, the existing matching detection methods for tunnel cracks have the problems of low matching efficiency and poor matching accuracy.

Images