Aiming at the problem that the current railway fastener detection algorithms are insufficient in practical applications

a detection algorithm for railway fastener based on regions of interest（ ROI） and polyline features is proposed. According to the stability feature of the background area in the fastener image

the ROI at the end of the fastener are extracted to overcome the interference of redundant information on the fastener detection. The distance field which can represent the fastener end structural information is obtained by the improved Euclidean distance transform in eight directions; the feature points of the polyline are computed by using the characteristic matrix which is obtained by fusing the distance fields corresponding to fastener bending direction

and the polyline features are extracted; defect detection is performed. Theoretical analysis and experimental results show that the proposed algorithm has the advantages of high accuracy

low complexity and strong robustness

and the missing rate and false detection rate reach 1. 3 % and 1. 7 % respectively

which verifies effectiveness of the algorithm.