Zipper anomaly detection method based on deep support vector data description model

The invention discloses a zipper anomaly detection method based on a deep support vector data description model. According to the scheme, the zipper anomaly detection method comprises the steps of collecting a zipper image; obtaining a training set, a test set and a verification set, and performing preprocessing; constructing an auto-encoder, and pre-training the auto-encoder by using the pre-processed training set zipper image block; taking encoder parameters in the trained auto-encoder as initial parameters of a feature extraction network in a deep support vector data description model, and using the pre-processed training set zipper image blocks to train thedeep support vector data description model; respectively inputting the zipper image blocks in the verification set and the test set into the trained depth support vector data description model to obtain an experience threshold T and an abnormal score s of the zipper image blocks in the test set, if s is greater than T, determining that the zipper blocks