This research is part of the development of the system, which aims to detect damage to railway tracks using the CLAHE-KNN model. This study aims to detect railway damage using the CLAHE-K-NN method and determine the effect of CLAHE on the K-Nearest Neighbor algorithm for railway damage detection. CLAHE is defined by Block Size (BS) and Clip Limit (CL). The dataset used in this study was railway imagery totaling 384 data. Based on the KNN method's experimental results, this method obtained an accuracy of 65.62%. In addition, the experimental results using KNN with CLAHE optimization get better accuracy by using clipLimit = 4.0 and tileGridSize = (4.8), which is 66.67%.