An Investigation into the Control of Brushed Yarn Properties: The Application of Machine Vision and Knowledge-based Systems. Part II: The Machine Vision System

Abstract

This paper describes the machine vision system which has been developed for real-time measurement of industrially brushed yarns. This system provides a means for determining the brushing effect and for continuously introducing appropriate corrective signals to the yarn brushing machine. The system measures the brushed yarn during production, enabling the acquired data to be analysed and a fast response to be made, thus ensuring uniformity of quality. The four steps in the inspection of a brushed yarn involved (i) image acquisition; (ii) image enhancement; (iii) feature extraction; and (iv) image analysis and interpretation are reported.