二维Fisher线性鉴别分析的图像分割算法,考虑了图像中目标和背景之间类间方差和类内方差在类别分离中的作用,有效地克服经典Otsu阈值法当图像中目标的面积很小(直方图上表现为峰的大小相差很大或者没有明显双峰)时产生的阈值"漂移"现象,是一种有效的图像分割方法.针对二维Fisher线性鉴别分析计算量大的特点,采用粒子群算法来搜索最优二维阈值向量.每个粒子代表一个可行的二维阈值,通过粒子群之间的协作来获得最优阈值向量.实验结果表明,所提出的方法不仅能准确地分割图像,而且计算量大大减少,达到了快速分割的目的,便于实时应用.%Two-dimensional (2-D) Fisher linear optimal discriminant analysis, which considers the gray information and spatial neighbor information between pixels in the image simultaneously, overcome especially if the histogram of images in reality has no distinct sharp valleys or the valley is flat and broad, the proposed is an efficient image segmentation method. However, finding the optimal threshold vector using exhaustive searching is expensive for 2-D fisher criterion function thresholding method. In this paper, an optimization method, i.e., particle swarm optimization (PSO) is used to find the optimal 2-D threshold vector, in which each particle represents a possible 2-D threshold vector and the best 2-D threshold is obtained through the cooperation among particles. To show the validity of the proposed method, this paper uses several infrared images to segment. Analysis and experimental results show that the proposed method can not only obtain ideal segmentation results but also decrease the computation cost reasonably, and it is suitable for real-time applications.
展开▼
