Active snake algorithm on the edge detection for gallstone ultrasound images

For the edge detecting of gallstone in ultrasound images, traditional approaches can only obtain discontinuous contours, due to the speckle noise and attenuation of sound wave. To solve the problem, we develop an active snake algorithm for detecting the edge of gallstone in ultrasound images. The proposed snake algorithm is active, it is always minimizing its energy function and therefore exhibits dynamic behavior. After initializing the snake curve, the snake along the edge locks onto the regions-of-interest (ROI), according to the position and radius of gallstone, then it shrinks the curve automatically to the desired shape, guided by image internal forces and external constraint forces. Experiments show that the proposed snake algorithm can detect the object edge more accurately and intelligently.