Self-adaptive detection method for overcut and gouges defects of workpiece

The invention discloses a self-adaptive detection method for overcut and gouges defects of a workpiece. The self-adaptive detection method comprises the following steps: S1, acquiring a sample data set of the workpiece; s2, acquiring a standard data set; s3, performing dynamic fitting on the standard data set by using a kth polynomial to obtain a detection function; s4, substituting the sample data set into the detection function to obtain a prediction parameter; s5, calculating a decision coefficient R2 between the sample data and the prediction parameters, judging the defect type of the sample data according to the decision coefficient R2, and if R2 is greater than or equal to 0.9, indicating that the sample data is a good workpiece; if 0.6 < = R2lt; 0.9, indicating that the sample data is an overcut defect workpiece; if R2lt; and 0.6, indicating that the sample data is a gouge defect workpiece. According to the invention, on the premise that the dynamic requirement of an industrial site is met and the scan