CV模型是一种重要的图像分割模型,本文针对其收敛速度慢、效率低的缺点提出一种求解CV模型的新方法.首先将CV模型的能量泛函改写成与原来有相同稳定解的总变分公式形式,然后使用对偶公式法求总变分公式的极小值,再在其中引入一速度项以加快模型的收敛速度.新方法一方面克服了梯度下降法要求时间步长小、迭代次数多的缺点,经过较少次的迭代就能收敛,减少了迭代计算的次数-CV model is one of the most important image segmentation model and disadvantage in view of its convergence rate is slow and low efficiency of the proposed a new method for solving the CV model. First, the energy functional of the CV model adapted and had the same stable solution total variational formula, and then use the dual formula method for total variation formula of minimum, and in which introduced a speed to accelerate the convergence speed of the model. New method to overcome the gradient descent method requires small time steps, the number of iterations of the shortcomings of the, after a few times of iteration will converge, reducing the number of iterative computation