Abstract: According to the existing manual methods to measure dimensional changes of fabric after laundering has an obvious error and low accuracy, especially the case with irregular shape, this paper proposes a method based on machine vision to calculate the fabric area change and then evaluate dimension stability of fabric after being washed. Six types of pure wool fabrics are chosen as experimental materials, which will be washed under different washing, rinse, spin time procedures before dry and image capture by machine vision system. The dimensional stability will be evaluated by taking the average dimensional change ratio not only in both warp and weft direction, but also in area before and after laundering, which will be calculated from the fabric images. Compared with the manual measurement, the computer vision method proposed in this paper can represent the dimensional change more globally and accurately, which provides a test basis for fabric design and washing program optimization.

Key words: fabric , dimensional stability , computer vision , washing parameter