Level Set Model driven by function of Signed Pressure Force For image segmentation
Messaouda Larbi, Rouini Abdelghani, Zoubeida Messali, Samira Larbi · 2019
Segmentation is an extremely active field of research since it represents one of the more difficult stages for relevant parameters extraction from images. In this work, we study. A New robust Level Set image segmentation model is developed by function of Signed Pressure Force (SPF). The advantages of this method are Firstly, the signed pressure function can effectively stop contours on weak or fuzzy edges. Secondly, the inner and outer limits can be detected regardless of the starting point of the initial contour. We compare this method with Chan Vese (C-V) method and Geodesic Active Contours (GAC). The performance of each method can be evaluated either visually, or from similarity measurements between the results of the segmentation and a reference. Through the results, we have shown that the better results are obtained with the proposed method.