hundunpso

  针对二维熵图像分割方法在求取最佳阈值时存在计算量大及微粒群算法容易陷 入局部最优且速度较慢等等问题, 提出了基于混沌粒子群优化算法的二维熵图像分割方法。 该方法考虑了图像中像素点灰度􀀂 􀀂 􀀂 邻域灰度均值对作为阈值对图像进行分割 利用混沌运 动随机性、遍历性和初值敏感性, 将混沌粒子群优化算法与阈值法相结合在二维空间作全局搜 索。实验结果表明了基于混沌粒子群优化算法的二维熵图像分割法用于阈值寻优减少了搜索 时间, 提高了收敛率。(Calculation of its large capacity and particle swarm algorithm is easy to fall to strike the best threshold for the two-dimensional entropy image segmentation method Into the local optimum and slower, the proposed two-dimensional entropy image segmentation method based on chaotic particle swarm optimization algorithm. The method takes into account the image pixel the gray 􀀂 􀀂 􀀂 neighborhood average gray value as a threshold for image segmentation chaotic transport Dynamic randomness, ergodicity and initial value sensitivity of chaotic particle swarm optimization algorithm with the threshold method in two-dimensional space for the global search On request. Experimental results show that the entropy method of image segmentation method based on chaotic particle swarm optimization algorithm for threshold optimization to reduce the search Time and improve the convergence rate.)

hundunpso.pdf,3815793,2012-02-26