ROBUST DETECTION AND CLASSIFICATION OF THE SPLICED YARN JOINT BY COMBINING LBG AND DTW

摘要

This paper presents an automatic vision based system for unsupervised detection and classification of spliced yarn joint. In the splice detection process, a competitive learning method based on LBG algorithm is used. In the splice classification process, a dynamic time warping (DTW) algorithm is used to classify the extracted splice joint into one of three degrees of quality based on the degree of similarity between the spliced joint and the non-spliced part of the same yarn. The use of DTW in the classification makes the proposed method adaptable to different types of yarns. Consequently, this method might be globally optimal for classification of all spliced yarn joint. The proposed method has been evaluated using three sorts of experiments showing a promising result.