Extracting Geometrical Features from a Continuous-filament Yarn by Image-processing Techniques

This paper presents a method of analysing the interstices of a section of continuous-filament yarn by applying image-processing techniques. To characterise the uncertainty in the shape of each interstice, the concept of connection degree and fuzzy surface based on fuzzy theory is introduced. The proposed connection degree can he determined from examples in order to be adapted to geometrical measurements and human impressions of the interstices analysed.