Evaluation of fabric smoothness based on photometric stereo images

This paper proposes a method for measuring fabric wrinkle using fabric 3-D surface re-modeling method based on photometric stereo algorithm. Firstly, according to illumination equation of an object surface, the relative surface orientation and gradient can be determined by introducing the conception of general inverse matrix, and surface depth can be obtained using curvilinear integral. Then the fitted surface can be further iterated and modified according to variational calculus and finite difference method, so the best re-modeling surface is obtained. Finally, applying it to the surface re-modeling of American Association of Textile Chemists and Colorists (AATCC) fabric smoothness replica, 3-D contour information of replica can be acquired and four smoothness feature parameters are furthered to extract to indicate smoothness degree. The results show that there is a good linear correlation between smoothness feature parameters and AATCC replica grades.