基于形状约束的快速图像分割方法

In this paper, a lattice Boltzmann method solving the energy functional with shape constraint is proposed for image segmentation. Firstly, the proposed algorithm aligns shape priors based on image moments. Secondly, the sparse shape priors in high dimensional space are projected into a low dimensional subspace via the locality preserving projections, and a shape-driven energy term is designed in the low dimensional space by a statistical method, i.e., kernel density estimation. Finally, the lattice Boltzmann evolution equation is deduced from the energy functional that is built by combining the shape-driven term with the data-driven term. The minimum value of the energy function is obtained by evolving of the LBM equation, and finally the image segmentation is accomplished. Compared with the traditional methods of solving the energy functional, this method could reduce the time consumption and improve the efficiency of image segmentation.