Textile quality detection algorithm and apparatus based on visual identification technology

Abstract

The invention discloses a textile quality detection algorithm and apparatus based on a visual identification technology, the textile quality detection algorithm based on the visual identification technology adopts a regional convolutional neural network, and the regional convolutional neural network comprises an input layer, a convolutional layer, a down-sampling layer, a full-connection layer and an output layer. During working, an image to be detected needs to be transmitted into the convolutional neural network. The image processing process comprises target candidate region generation, defect feature extraction, defect specific identification and positioning and optimal defect position determination, the defect detection based on the regional convolutional neural network is an end-to-end detection model, image preprocessing is not needed, the defect position can be accurately given while the defect is identified, and the invention has the characteristics of high efficiency and accuracy.

Description

technical field [0001] The invention relates to the technical field of visual recognition, in particular to a textile quality detection algorithm and equipment based on visual recognition technology. Background technique [0002] With the rapid development of industrialization in our country, the production speed of various industrial products has been greatly improved. Due to the various types of defects in industrial products, the defect identification of products in traditional industries has always been done manually, but manual identification Defects will be affected by many factors, and the missed detection rate is very high, which has a negative impact on product quality and corporate reputation. Therefore, the present invention provides a textile quality detection algorithm and equipment based on visual recognition technology, which can automatically learn and identify multiple defects. Such defects, high accuracy, save a lot of manpower and material resources. Con...

AI Technical Summary

Problems solved by technology

[0002] With the rapid development of industrialization in our country, the production speed of various industrial products has been greatly improved. Due to the various types of defects in industrial products, the defect identification of products in traditional industries has always been done manually, but manual identification Defects will be affected by many factors, and the missed detection rate is very high, which has a negative impact on product quality and corporate reputation. Therefore, the present invention provides a textile quality detection algorithm and equipment based on visual recognition technology, which can automatically learn and identify multiple defects. defects, high accuracy, saving a lot of manpower and material resources

Images