Textile fabric fluff detection method based on deep neural network

• Main Author: ZHENG ZHENBIN
• Format: Patent
• Language: Chinese; English
• Published: 16.03.2021
• Subjects: CALCULATING; COMPUTING; COUNTING; HANDLING RECORD CARRIERS; IMAGE DATA PROCESSING OR GENERATION, IN GENERAL; PHYSICS; PRESENTATION OF DATA; RECOGNITION OF DATA; RECORD CARRIERS

The invention discloses a textile fabric fluff detection method based on a deep neural network, and the method comprises the steps: obtaining a to-be-detected textile fabric image, wherein the to-be-detected textile fabric is a textile fabric in production; calculating the difference between the image of the to-be-detected textile fabric and the image of the textile fabric which is detected to bequalified on the texture level so as to obtain a texture difference feature map; inputting the texture difference feature map into a convolutional neural network to obtain a first feature map; equallydividing the first feature map into a plurality of image blocks according to different scales, wherein the sum of the number of the image blocks of the different scales is equal to the number of nodes of the input end of a full connection layer; and enabling the plurality of image blocks to pass through the full connection layer and the classification function to obtain a classification result, wherein the classification