Improved Level Set Model for Color Image Segmentation
Lei Wang, Sheng Wang, Yi Liao, Xiufen Ye, Tian Wang · 2018
Color image segmentation has been widely applied to diverse fields in the past decades for containing more information than gray ones. The traditional level set algorithm calculates the level set energy function based on the gray value of the gray image. A level set algorithm based on color space is proposed, which is based on weightings of different channels, and the energy function of the level set is rewritten. The k-means algorithm is used to pre-segment, initialize the level set contour curve and the sign distance function, and achieve the unsupervised segmentation of the algorithm. The steps of the algorithm are given. The adaptability and accuracy of different types of color image are solved. Compared to other popular algorithms, it has the competitive performances both on speed and accuracy. The experiments performed on real-world data sets demonstrate the validity of the proposed algorithm.