Tunnel surface two-dimensional laser image enhancement processing method

Abstract

The invention relates to a tunnel surface two-dimensional laser image enhancement processing method. The method comprises the following steps of S1, collecting a to-be-detected tunnel surface two-dimensional laser image; S2, determining an exposure elimination algorithm threshold value according to the pixel value characteristics of the two-dimensional laser image on the surface of the tunnel to be detected; S3, performing histogram equalization processing on the two-dimensional laser image of the surface of the tunnel to be measured to obtain an equalized two-dimensional laser image; S4, enhancing the equalized two-dimensional laser image target features according to a Laplace enhancement algorithm, eliminating image noises through Gaussian filtering, and acquiring a processed two-dimensional laser image; S5, recognizing image splicing seam position information according to the processed two-dimensional laser image; S6, according to the image splicing seam position information, adopting a gray value processing algorithm to eliminate a splicing seam, and acquiring an enhanced two-dimensional laser image. According to the invention, the coordination of the high pixel domain and thelow pixel of the image can be effectively improved, and the detail information effect of the tunnel image is enhanced.

Description

technical field [0001] The invention relates to the field of image processing, in particular to a two-dimensional laser image enhancement processing method on the surface of a tunnel. Background technique [0002] At present, image enhancement algorithms are divided into two categories: spatial domain method and frequency domain method. The spatial domain method adjusts the central pixel by analyzing the influence of local pixels on the central pixel. The representative algorithm has histogram equalization. The advantage of histogram equalization is that it can display better details in overexposed or underexposed photos. The disadvantage is that it does not select the processed data, which may increase the contrast of background noise. Thereby reducing the contrast of the useful signal, reducing the transformed gray level, and some information disappears. The frequency domain method is to modify the transformation coefficient of the image through certain rules, and then pe...

AI Technical Summary

Problems solved by technology

The Laplace transform can get the outline of the image, so as to enhance the image. The disadvantage is that it will generate noise during the change process, which will affect the quality of the image.

At present, the research on tunnel images mainly focuses on the detection and identification of tunnel defects, but there are relatively few studies on tunnel image enhancement.

Images