Railway sleeper defect detection method based on terahertz imaging

The invention discloses a railroad sleeper defect detection method based on terahertz imaging, which comprises the following steps of: scanning a railroad sleeper to be detected by using a terahertz imaging device to generate a grey-scale map; preprocessing the grey-scale map to obtain a processed grey-scale map; performing sobel edge detection on the preprocessed grey-scale map to obtain featurepoints of linear distribution of sleeper boundary lines; introducing the characteristic points of sleeper boundary line linear distribution into Haff linear detection, and extracting a sleeper area tobe subjected to defect detection; and presetting a gray value matrix, comparing the sleeper area to be subjected to defect detection with the gray value matrix, and obtaining a sleeper defect judgment result through sleeper defect identification based on gray correlation. The track defect detection method can solve the problems of low efficiency, high strength and insufficient safety of a track defect detection method in