Method and system for automatic identification of pavement disease images

Abstract

The invention discloses an automatic identification method and system for road surface disease images. The method includes: preprocessing the captured road surface image, including Gamma grayscale correction, Gaussian filter enhancement and local adaptive binarization; performing edge detection on the binarized image; performing connected domain contour detection on the image after edge detection , to obtain the number of connected domains and the circumscribed rectangle of the connected domain outline; judge the area where the road surface crack is located according to the shape of the circumscribed rectangle of the connected domain outline; extract the image of the crack area from the image after edge detection according to the location information of the area where the pavement crack is located , and superimposed the black template to form a crack feature image with the same size as the original image; the crack feature image is classified based on the convolutional neural network. Compared with the prior art, the present invention has very excellent detection efficiency for crack location in pavement disease, and has good robustness performance for crack images with different characteristics.

Description

technical field [0001] The invention relates to an automatic identification method and system for road surface disease images, belonging to the technical field of image processing. Background technique [0002] Among pavement diseases, pavement cracks are the main manifestation of pavement damage in high-grade highways, and are very important for modern and efficient highway maintenance. Because the traditional manual detection methods are time-consuming, inaccurate, highly dangerous, obstruct traffic, and have large subjective differences, high-precision cameras are now used to quickly capture road images for automatic computer detection. Various pavement crack detection and localization algorithms have been proposed. [0003] According to the image characteristics of the low gray value of the crack area, Xie Changrong et al. published the research on the image processing algorithm of pavement crack detection by analyzing the excellence of various classic image processing ...

AI Technical Summary

Problems solved by technology

However, for complex situations that contain more non-crack information and more interference objects, the ideal recognition effect cannot be achieved.

Images