Abstract：Abstract: In order to enhance the robustness to initial contour as well as improve the segmentation efficiency for images with intensity inhomogeneity or noise. Therefore, a novel region-based active contour model was proposed. First, a global intensity fitting term and a local term were designed separately, the model’s fitting term was obtained by linear combination. The weight of these two terms were adjusted to improve the robustness of the model to contour initialization. Finally, a length term was employed to keep the smoothness of evolving curve. Experimental results show that compared with RSF model and SLGS model , the proposed model has the number of iterations reduced by about 60% and 40%, and the segmentation time reduced by about 70% and 10%. The proposed model can quickly and accurately segment noisy images and images with intensity inhomogeneity without initial contour. Besides, it can be applied on some medical images and infrared images.
