传感器图像配准是传感器融合中的一个重要课题 ,它的任务就是在不同的传感器图像之间找出正确的映射匹配关系。从优化问题的角度考察传感器配准问题 ,提出一个适当的适配度函数以评定传感器图像间的映射参数 ,并在此基础上采用遗传算法、模拟退火、遗传模拟退火算法等随机优化方法解决此优化问题 ,从而达到自动配准传感器图像的目的。实验结果证明了这种配准方法的可行性 ,而且它对噪声不敏感 ,表现出较好的鲁棒性。遗传模拟退火算法更是在相对较少的迭代后即获收敛 Sensor image registration is an important issue in sensor fusion. Its task is to find the correct mapping match between different sensor images. The problem of sensor registration is investigated from the perspective of optimization problems. An appropriate fitness function is proposed to evaluate the mapping parameters between sensor images. Based on this, a stochastic optimization method such as genetic algorithm, simulated annealing and genetic simulated annealing algorithm is used to solve the problem Optimize the problem, so as to achieve the purpose of automatically registering sensor images. Experimental results show the feasibility of this registration method, and it is insensitive to noise and shows better robustness. Genetic simulated annealing algorithm converges even after relatively few iterations