Adhesive tape damage detection method and system based on support vector machine

The invention relates to an adhesive tape damage detection method based on a support vector machine, which comprises the following steps of: S1, acquiring photoelectric data of a sample containing a damage mark, and preprocessing the photoelectric data; s2, wavelet transform is adopted to obtain the characteristics of the photoelectric signals, and a data set is constructed; s3, training a support vector machine according to the obtained data set to obtain an adhesive tape damage detection model; s4, a photoelectric sensor obtains photoelectric signals on a production line in real time, the photoelectric signals are preprocessed and then input into the adhesive tape damage detection model, and an identification result is obtained; and S5, controlling an execution module to screen out the damaged adhesive tape according to an identification result. The photoelectric sensor and the support vector machine are combined for damage detection, the efficiency and the accuracy of damaged adhesive tape identification c