Abstract: To tackle the difficulty of extracting tongue crack using computer technology in the researches on modernizing tongue diagnosis of traditional Chinese Medicine,this paper proposed a novel algorithm for the extraction of tongue cracks.First,the false-color image is enhanced for the color fissured tongue image using the kernel false-color transformation.Then,the gradients are computed with the second component of the false-color image in RGB color space.Finally,the gradient image is binarized by a thresholding technique,such as the hysteresis threshold.Experimental results show that the proposed method can effectively isolate the tongue cracks from the surrounding tongue proper and tongue coat,and accurately segment the tongue cracks.The proposed algorithm offers a credible means for the computerized researches of tongue crack.