The automated segmentation of the tongue body is a premise to establish an automatic diagnosis system according to the features of tongue in traditional Chinese medicine(TCM),whose qualities have great effect on the performance of tongue diagnosis.However,automated tongue segmentation is difficult due to the complexity of pathological tongue,variance of tongue shape and interference of the lips.Here a novel algorithm for automated tongue segmentation was presented based on hyperspectral tongue image data acquired from a hyperspectral imaging system.First,by finding the spectral angle(SA) between each pixel and every other pixel in the original data cube,a transformed data cube was constructed.Thus,each spectrum in the transformed SA cube contained information about spatial changes in the tongue scene.Then,each spectrum in the transformed SA cube was analyzed with a one-dimensional edge-detector.Finally, the whole contour of the tongue was extracted from the hyperspectral tongue image according to the edge detected.Experimental results demonstrate that the novel tongue segmentation algorithm can segment the tongue more accurately.