c语言实现灰度图像阈值分割,几种图像阈值分割算法的实现与比较

摘要：图像分割是进行图像分析的关键步骤，也是进一步理解图像的基础。该文主要论述了常用的几种图像阈值分割的算法及原理，并以研究沥青混合料的集料特征为背景，从实验角度对图像阈值分割的直方图阈值法、迭代法和大津法进行了分析比较，得出了结论。

关键词：图像分割；直方图阈值法；迭代法；大津法

中图分类号：TP391 文献标识码：A文章编号：1009-3044(2011)13-3109-03

Achieve and Comparison of Image Segmentation Thresholding Method

CHEN Ning-ning

(Department of Technology, Xi'an International University, Xi'an 710077, China)

Abstract: Image segmentation is a key step for image analysis, Is also the basis for further understanding of the image. In this paper, discusses several commonly used image segmentation algorithms and theory, and to study the aggregate asphalt mixture characteristics of the background, experimental results are shown to compare histogram threshold, Iteration method and the Otsu.

Key words: image segmentation; histogram threshold; iteration method; Otsu

1 概述

图像分割是进行图像分析的关键步骤，也是进一步理解图像的基础。图像分割是指将图像划分成若干个互不相交的小区域的过程，小区域是某种意义下具有共同属性像素的连通集合[