Detection of tunnel water leakage based on image processing

In this paper, a novel method of tunnel water leakage detection and recognition method is proposed. It combines the analysis of leakage intensity feature and the application of Artificial Neural Networks (ANNs) algorithm. Firstly, the original color image is transformed into a gray value image. Then, the canny filter is used to extract the ragged edge. After that, the non-maximum-suppression method is adopted to remove the noise and hysteresis threshold is utilized to obtain the accurate edge. An experiment was conducted to test the accuracy and stability of the proposed method.