摘要：作为图像处理的关键技术之一，图像分割在诸多领域都具有广泛应用，因而成为近年来所被大量学者研究的一门热门课题。本课题主要探讨了一些常用的分割算法，分析其各自实现原理，实现步骤以及所具备的优缺点。并针对课题要求，对模糊C均值聚类算法，Meanshift算法以及Otsu阈值算法进行编程设计，最终在Matlab环境下分别对这三个算法进行仿真。之后通过对算法的研究和主观的观测对仿真结果进行对比，可以发现，相对于Otsu算法，模糊C均值聚类算法能够使图像细节得以更好的保留，而Meanshift算法则相对于模糊C均值聚类算法有更好的抗噪性能。通过分析这些算法各自的优缺点以及适用性，能更好地为我们在不同场景下对分割算法的选取提供依据。
关键词 图像分割 自适应阈值 模糊C均值聚类
毕业设计说明书外文摘要
Title Research on Image Segmentation Algorithm Based on Clustering
Abstract：As one of the key technologies of image processing, image segmentation has been widely used in many fields, which has become a hot topic in recent years by a large number of scholars.This paper mainly discusses some commonly used segmentation algorithms, analyzes their respective implementation principles, implementation steps, advantages and disadvantages. For the subject requirements, the fuzzy C-means clustering algorithm, the Meanshift algorithm and the Otsu threshold algorithm are programmed and simulated in the Matlab environment finally.Then, the simulation results are compared by the research of the algorithm and the subjective observation, it is found that the fuzzy C-means clustering algorithm can preserve the details of the image better than the Otsu algorithm, and the Meanshift algorithm has better anti-noise performance than the fuzzy C-means clustering algorithm. By analyzing the advantages and disadvantages of these algorithms and their applicability, we can provide the better basis for the selection of the segmentation algorithm in different scenarios.
Keywords Image segmentation Clustering Fuzzy C - means clustering
目 次
1 绪论 2
1.1 研究背景及意义 2
1. 3 研究内容及章节安排 3
2 常用的图像分割方法 5
2.1 基于阈值的分割方法 5
2.2 基于边缘的分割方法 7
2.3 基于区域的分割方法 8
2.4 基于数学形态的分割方法 9
2.5 基于特定理论的分割方法 9
3 聚类分析 11
3.1 聚类分析综述 11
3.2 K-means聚类算法 12
3.2.1 K-means聚类算法原理 12
3.2.2 K-means聚类算法具体步骤 12
3.2.2 K-means聚类算法优缺点 13
4 基于模糊聚类图像分割 15
4.1 模糊C均值聚类算法 15
4.1.1 算法原理概述 15
4.1.2 算法步骤 16
4.2 基于Meanshift的模糊C均值聚类算法 16
4.2.1 算法的改进背景 16
4.2.2 算法原理概述 17
4.2.3 算法步骤 17
5 实验及分析 19
5.1 模糊C均值聚类算法图像分割 19
5.2 Meanshift聚类算法图像分割 20
5.3 otsu阈值算法图像分割 22
5.4 仿真结果分析 23 基于聚类的图像分割算法研究:http://www.chuibin.com/tongxin/lunwen_205978.html