摘要
图论是一种新的图像分割算法,近几十年来,一直是国内外的研究热点.但当图像很大时图论分割算法很耗时.针对这个缺点,提出了一种基于阈值和图论的图像分割算法.该算法先对图像进行平滑、锐化等预处理,然后再用最大方差法把目标和背景分隔开,最后对目标图像应用图论最小生成树(MST)的方法进行再分割.通过一系列的图像试验,该算法能够较准确地分割图像.与传统的图论分割算法相比,该算法对大部分图像有较好的分割效果.
Graph theory is a new image segmentation algorithm, which has been a hot research topic both at home and abroad in recent decades. But when the image is very large, it is time consuming to use the graph theory. In view of this shortcoming, this paper proposes an image segmentation algorithm based on threshold and graph theory, which firstly do the image smoothing, sharpening and other pretreatment; then the maximum variance method is used to separate the target and the background; finally, the spanning tree minimum（MST）method is used to resegment the target image using graph theory. Through a series of image experiments, the algorithm can accurately segment the image. Compared with the traditional graph theory segmentation algorithm, this algorithm has a good segmentation effect on most of the images.
出处
《宁德师范学院学报（自然科学版）》
2016年第1期62-66,共5页
Journal of Ningde Normal University(Natural Science)
关键词
图论
阈值
最大方差阈值
图像分割
graph theory
threshold
maximum variance threshold
image segmentation
