基于MATLAB的图像分割技术

摘要一幅图像中所包含的信息量可能十分巨大，全部处理起来肯定效率很低，因此对图像分割的研究就显得尤为重要。本文主要研究了基于阈值检测、边缘检测以及区域检测算法的图像分割技术，并对其经典算法进行了MATLAB仿真。采用峰值信噪比对各种方法进行了评价，对比分析了各种算法的优缺点和适用范围，以及不同类型的图像分割算法的适用条件。结果表明，四叉树算法处理后的图像，在边缘的精度上比上述所有分割方法都更为优秀，而且对图像中噪声的抗干扰能力也十分出色。27278
关键词图像分割阈值检测边缘检测区域检测 MATLAB 四叉树
毕业论文设计说明书外文摘要
Title    The Image Segmentation Technology Based on MATLAB
Abstract
An image can cover very large amount of information, which is difficult to handle all with efficiency, so，the study of image segmentation is particularly important.This article mainly studied based on threshold detection, edge detection and region detection algorithm of image segmentation technology, and performed MATLAB simulation based on their calssical algorithms. These methods was evaluated by peak signal-to-noise ratio, and their advantages and disadvantanges were compared, applicable scope were analyzed，as well as the suitable conditions of diffierent types of image segmentation alogrithm.Results show that the processed image quadtree algorithm, on the precision of the edge is more excellent than all of the above method, and the anti-jamming ability of noise in the image is also very good.
Keywords   image segmentation threshold detection edge detection MATLAB   region detecetion classical algorithm quadtree
目   次
1 绪论   1
1.1 图像分割的意义   1
1.2 图像分割技术的现状和发展趋势   1
1.3 本文主要内容   2
2 图像分割算法研究   3
2.1 基于阈值的图像分割法   3
2.1.1 双峰法   4
2.1.2 迭代法   5
2.2 边缘检测   6
2.2.1 Roberts算子   9
2.2.2 Prewitt算子   9
2.2.3 Sobel算子   10
2.2.4 LOG（Laplacian-Gauss）算子   10
2.2.5 Canny算子   11
2.3 区域分割法   13
2.3.1 区域生长法   13
2.3.2 区域分裂合并   14
3 MATLAB仿真结果及评价   17
3.1 基于MATLAB的图像分割技术   17
3.1.1 MATLAB简介   17
3.1.2 图像分割技术原理   17
3.2 阈值分割法仿真   18
3.2.1 双峰法阈值分割仿真   18
3.2.2 迭代法阈值分割仿真   20
3.3 边缘检测仿真   21
3.3.1 Roberts算子边缘检测的仿真   22
3.3.2   Prewitt算子边缘检测的仿真   23
3.3.3 Sobel算子边缘检测的仿真   23
3.3.4   LOG算子边缘检测的仿真   24
3.3.5 Canny算子边缘检测的仿真   26
3.3.6 边缘检测算子分割结果分析   27
3.4 区域分割算法仿真   29
3.4.1 区域生长算法边缘检测的仿真   29
3.4.2 区域分裂合并算法边缘检测的仿真   30 基于MATLAB的图像分割技术:http://www.751com.cn/tongxin/lunwen_21736.html