Based on improved local feature forming failure and mesh defect detection

In this paper, an improved method for detecting poor local feature forming and mesh defects is designed, which includes: acquiring the cake image, image downsampling; Image segmentation and image cropping; Affine transformation; Get the data set; The texture feature vectors of each image in the training set are calculated by the extended LBP algorithm. Generate XML file; The texture feature vectors of each image in the test set were calculated. The SVM support vector machine predicted the label of each image in the positive and negative samples of the test set, and the prediction accuracy was calculated.[1] Slider prediction was performed on the detected cake images. Judge the result according to the label output predicted by the slider. The invention removes the final background interference through image segmentation, so that the processed image no longer contains a large amount of interference, and solves the problem from staying in the laboratory research and verification stage to being directly used in the industrial production line. The invention has high accuracy and short time for detecting the defects of textile silk cake forming.