关键词: 混合测地区域活动轮廓模型, 柔化核函数, 符号压力函数, 局部信息熵, 非凸正则化项

Abstract: Aiming at the phenomenon that the variational level set algorithm takes large amount of calculation and converges slowly in the process of image segmentation, this paper proposes a new hybrid geodesic region active contour model of local entropy based on the predecessors’ research. The model constructs a new energy functional, introduces a mollifying kernel function to be window function, constructs a new signed pressure force function to replace the geodesic edge stopping function, and uses local entropy as the weight of image fitting energy, then adds a nonconvex regularization term to constrain the level set function. The algorithm getting from this not only accelerates the convergence rate of the contour curve, but also can address image segmentation inaccuracy for image intensity inhomogeneity or blurring caused by the change of illumination or other external factors. The simulation experiment results on synthetic images and real images show that the proposed algorithm is of higher converging speed and better accuracy, less sensitive to the location of initial contour at the same time, and has better robustness.

Key words: hybrid geodesic region active contour model, mollifying kernel function, signed pressure force function, local entropy, nonconvex regularization term