Grading of Yarn Appearance Using Image Analysis and an Artificial Intelligence Technique

Abstract

In this research, a new method is used for grading of yarn appearance based on yarn images of ASTM standard (section D 2255), by using an image processing technique and an artificial intelligence technique. In this method, grading of yarn appearance is based on computer vision and analyzing the images of standard pictorial boards of yarn. Therefore this method is very similar to human vision. The logic of the classification by ASTM is considered and then a new definition for classification of yarn appearance grade is presented. In this method of classification, the grading procedure is not dependent on yarn structure and raw materials. Thus it is possible to use this method for grading of any type of yarn based on apparent features.