混沌微粒群优化算法在图像匹配中的应用

针对传统图像匹配计算量较大、匹配速度慢、抗干扰能力差的问题，将混沌算子与微粒群优化算法相结合，提出一种鲁棒性强、计算速度快的图像匹配方法。该算法利用微粒群优化算法的收敛快速性和混沌运动的遍历性、随机性等特点，实现了非遍历性搜索。在算法初始化阶段，对粒子位置混沌初始化；在算法运行期间，对优秀个体进行混沌扰动避免落入局部最优。提高了算法对多维空间的全局搜索能力，并可以有效避免早熟现象。实验结果表明该算法的图像匹配具有快速性和较高的准确性，对解决噪声情况下的图像匹配问题十分有效。

Application of Chaos Particle Swarm Optimization algorithm in image matching

For the problems of the image matching are computationally expensive and slow speed and poor robustness,by introducing chaos state into the original Particle Swarm Optimization(PSO),this paper proposes a new algorithm Chaos Particle Swarm Optimization (CPSO).The new algorithm makes good use of the properties-ergodicity,randomicity, and initial sensitivity of chaos, which realizes non-ergodic searching and can be used to find the best matching point very quickly.At the beginning,the location of the particle is evaluated by chaos.During the running time,chaos perturbation is utilized to avoid the search being trapped in local optimum.CPSO is able to search the global optimizer and avoid the premature convergence on the multidimensional variable space.The experimental results indicate that this approach has high speed and accuracy in image matching and is very effectivefor image matching processing with noise.