医学图像的亚像素点边缘检测.doc

医学图像的亚像素点边缘检测

1.84万字
自己原创的毕业论文，仅在本站独家出售，重复率低，推荐下载使用

摘要 医学图像的边缘检测是图像分割、目标识别、区域形状提取等图像分析领域十分重要的基础。在进行医学图像理解和分析时，第一步往往都是边缘检测。目前，边缘检测已成为机器视觉研究领域最活跃的课题之一，其研究具有非常重要的理论意义和实际应用价值。传统的边缘检测方法的检测精度最高只能达到一个像素级，但是，随着科学技术的飞速发展，检测等应用对精确度的要求不断提高，传统的像素级边缘检测方法已经不能满足实际测量的需要，因此需要更高精度的边缘检测方法，即亚像素边缘检测方法。本文对图像的亚像素边缘检测进行了深入研究，包括以下几个方面：（1）本文详细阐述了数字图像边缘检测技术和亚像素边缘检测技术在国内外取得的研究成果和发展现状，论述了亚像素边缘检测技术的研究意义，同时，分析了亚像素边缘检测技术在图像处理领域未来的发展趋势。（2）对几种经典的像素级边缘检测算法进行了研究，主要包括 Roberts 算子、Sobel 算子、Prewitt 算子、Laplacian边缘检测算子， LOG算子， Canny算子等，并使用 Matlab 2010对这几种算法进行仿真实验，对比了他们的优缺点及适用范围。（3）对几种常用的亚像素边缘检测方法进行了深入研究，包括插值法，拟合法，矩法。（4）对二次曲线拟合方法做了重点研究，先用Canny算子得到像素级边缘，然后通过二次曲线拟合亚像素边缘提取公式确定出图像边缘。最后，使用Matlab 2010实现本文提出的曲线拟合亚像素边缘检测方法，实验结果表明：本文提出的方法能够使检测出的图像边缘达到亚像素精度。

关键词 边缘检测 亚像素边缘检测 曲线拟合


Sub Pixel Edge Detection on Medical Image
Abstract Edge detection of the medical image is a very important basis in image analysis fields of image segmentation, target recognition, region shape extraction, etc. In image understanding and analyzing, the first step is always edge detection. At present, the edge detection has become one of the most active subjects in machine vision research fields, which has very important theory significance and practical application value. The detection accuracy of the traditional edge detection algorithms can only reach a pixel level. However, with the rapid development of science and technology, industrial detection and other applications for accuracy requirements is increasing. The traditional pixel level edge detection algorithms have been unable to satisfy the need of practical measurement, so we need the edge detection algorithms with higher accuracy, namely the sub pixel edge detection algorithm. In this paper, the sub pixel edge detection technology is studied deeply, and the research contents include the following several aspects: 1）In this thesis, the research achievements, the present development situation of the digital image edge detection technology and the sub pixel edge detection technology at home and abroad is described. And the research significance of the sub pixel edge detection technology is discussed. At the same time, the future development trend of the sub pixel edge detection technology in image processing field is analyzed. （2）Several classic pixel level edge detection algorithms are studied, which mainly include Roberts、 Sobel、 Prewitt、 Laplace、 LOG and Canny operator, etc. And several algorithms are simulated with MATLAB 2010. Their advantages, disadvantages and applicable scope have been Compared. (3）Several sub pixel edge detection algorithms which are commonly used are studied, including the interpolation algorithm, the curve fitting algorithm, and the moment algorithm, etc. (4) the quadratic curve fitting algorithm is studied intensively . firstly, we use the Canny operator to get the pixel level edge,and then through the quadratic curve fitting to extract the edge of the image .Finally, the algorithm has been implemented with the MATLAB 2010. The experimental results show that we can detect the edge of the digital images at sub-pixel accuracy by the proposed algorithm.
Keywords  Edge detection Sub pixel edge detection Curve fitting


目录
第一章 绪论 1
1.1 课题背景 1
1.2 国内外发展现状 2
1.3 边缘检测的发展趋势 4
1.4 课题的研究目的及意义 5
第二章 边缘检测算子 6
2.1 边缘检测技术概述 6
2.2 像素级边缘检测算子 6
2.2.1 Roberts边缘检测算子 6
2.2.2 Sobel边缘检测算子 7
2.2.3 Prewitt边缘检测算子 7
2.2.4 Laplacian边缘检测算子 8
2.2.5 LOG算子 8
2.2.6 Canny边缘检测算子 9
2.2.7 实验结果及分析 10
第三章 亚像素边缘检测方法 13
3.1 亚像素定位原理 13
3.2 常用的亚像素边缘检测算法 14
3.2.1 插值法 14
3.2.2 拟合法 19
3.2.3 矩法 24
第四章 基于二次曲线拟合的亚像素边缘检测方法 29
4.1 二次拟合法的理论基础 29
4.2 算法流程 31
4.3 实验结果 32
发展与展望 34
致谢 35
参考文献 36