Color separation algorithm for mixed dyed textiles based on image segmentation

The computer image segmentation technology is used to process the scanned high resolution images of mixed dyed textiles,and an image segmentation method is presented based on the color characteristics of the containing fibers.Firstly,the color image is turned into the HSV color space,and then the gray region and color region of the image are segmented as component V and component H separately with Fuzzy C-means clustering method(FCM).The test results proved that this method can separate the image into several useful kinds,retaining the color and texture of the original image.The algorithm is suitable for computer aided color mixing for textile,and is more direct and efficient than the traditional forecast method based on optical models.