马尔可夫随机场中应用蚁群系统的红外图像分割

利用蚁群系统(ACS)的信息正反馈和启发式的搜索特点,针对马尔可夫随机场(MRF)图像模型的局部相关特性和最大后验概率(MAP)估计,可以在较短时间内得到图像分割目标函数的全局最优解,从而可以避免传统模拟退火算法的庞大时间复杂度.通过对军用红外图像的分割实验,可以看出这种算法能够在抗噪声、保留目标边缘和降低时间复杂度方面得到较满意的结果.

Markov Random Field with Ant Colony System for IR Image Segmentation

Information positive feedback and heuristic search,the characters of Ant Colony System(ACS),were applied for the image segmentations in this paper.The MAP global best solution of segmentations will be got though Markov Random Field(MRF),which describes image data relations by local correlations instead of global image possibility distributions.Compared with the Simulated Annealing(SA),ACS needs less time to search the global best solution.The followed segmentations experiments of IR military images proved that this novel algorithm could reach a satisfied result among the noise restraint,edges preservation and time complexity.