Sequential image for measurement of fabric crease recovery angle

Abstract

Crease recovery, one of an essential property for assessing fabric usability, is often evaluated by crease recovery angle method. This paper presents an image registration method based on mutual information to measure the crease recovery angle. The mutual information of two random variables is a measure of their mutual dependence. The maximum normalized mutual information can be worked out when the positions of the free wing in both the rotated frame and the other frame achieve a perfect match. The difference of recovery angles between two frames is defined as the rotated angle of the former frame when the normalized mutual information of the rotated frame and the latter frame is the maximum. Compared with Hough transform method, the recovery angle measured by the proposed method is more accurate when the free wing bends. Besides, mutual information is also applied to determine the stable time when the recovery process reaches a steady state by measuring it between a frame and the other frame which is captured 10 s after the former frame. The experimental results show that the crease recovery properties of wool–polyester-blended fabrics are better than the fabrics made of other materials. However, the property of 100% cotton fabric is poorer among the selected fabrics. It is demonstrated that the method has excellent robustness and adaptability in recovery angle calculation.