改进模拟退火算法在图像配准中的研究

为了实现图像的配准，提卅了使用模拟退火算法求解2幅图像的最佳匹配参数，然后对待配准图像进行变换，从而达到配准目的．实验表明，该方法对平移、旋转的2幅图像具有较高的配准精度和初值鲁棒性，其中的改进算法，即单纯形一模拟退火算法可以使优化解不陷入局部极值而获得全局优化解，具有更高的配准精度，同时也大大提高了运算效率．

Research of improved simulated annealing algorithm in image registration

To achieve image registration, a simulated annealing algorithm is presented to search the optimal matched parameters of two images, and the registration could be achieved through transformation to the registrated images. Experimental results show that the proposed approach can efficiently achieve high registration precision and initial value robustness for two translated and rotated images. The new improved method, or simplex-simulated annealing algorithm, can prevent the optimizing process from being trapped into local extremum so that the global optimal solution is obtained. This algorithm has higher registration precision , and the computation efficiency can be improved substantially.