Detection of foreign fibers among cotton based on image multi-resolution difference method

It is difficult to detect the non-homochromo foreign fiber among cotton fibers. For this reason, a method is proposed for detection of foreign fiber based on the image multi-resolution difference. First of all,different frequency components of a detection image are separated through the wavelet decomposition. Then,different resolution sub-graphs are used for differential seperation to improve information contrast of foreign fiber and cotton. On this basis,the maximum between-cluster variance method is proposed for binarization segmentation of the multi-resolution difference sub-graph to obtain the foreign fiber information. Finally,the information fusing of the segmented sub-graphs is conducted so that the detection of non-homochromo foreign fiber is realized. Experimental results show that the method can effectively restrain the cotton background information,extract foreign fiber information to the maximum extent,and achieve a detection accuracy of above 90% for non-homochromo foreign fiber.