Image Segmentation of CV Model Based on Curve Area Constraint

Traditional CV models do not take into account the area energy terms inside and outside the evolution curve for active contour segmentation, This slows down the convergence rate of the evolution curve. In this paper, the energy function in the curve is constructed for the CV model problems. An edge function is constructed by using Gauss convolution kernel function, And the edge energy is used to construct the area energy in the evolution curve, the convergence rate of evolution curve is improved by the method proposed in this paper. The experimental results show that, the model proposed in this paper can automatically detect the inner contour of perforated targets, it is robust to the shape and position of the initial contour, and it can quickly calculate the global minimum value of the image. For single target and multi target images, The method of this paper can quickly and accurately segment the image. Compared with the traditional Chan-Vese model, the LBF model and the LIF model, Using the method of this paper to segment images satisfies real-time requirements.