|
|||||
|
|
| 新的噪声污染灰度图像边缘检测方法 | |
| 引用本文： | 陈红艳,马上,王海江.新的噪声污染灰度图像边缘检测方法[J].计算机工程与应用,2010,46(11):183-185. |
| 作者姓名： | 陈红艳 马上 王海江 |
| 作者单位： | 1. 西南科技大学信息工程学院,四川,绵阳,621010 2. 电子科技大学通信抗干扰国防重点实验室,成都,610054 3. 电子科技大学电子工程学院,成都,610054 |
| 基金项目： | 西南科技大学青年预研基金No.07zx3112~~ |
| 摘 要： | 传统的图像边缘检测方法由于引入了各种微分运算，因此用于噪声图像边缘检测时对噪声极度敏感。针对这一问题，提出了一种基于独立分量分析技术的噪声图像边缘检测方法，该算法通过计算数据之间的高阶统计信息，提取特征模板，然后将被高斯噪声污染的灰度图像与这些模板逐个匹配，提取出边缘成分。实验结果表明，基于独立分量分析技术的模板匹配方法自适应强，复杂度低，是一种有效的高斯噪声污染灰度图像边缘检测方法。 |
| 关 键 词： | 边缘检测 噪声图像 独立分量分析 特征模板 |
| 收稿时间： | 2008-10-7 |
| 修稿时间： | 2008-12-25 |
New method for edge detection of noisy gray scale image |
|
| CHEN Hong-yan,MA Shang,WANG Hai-jiang.College of Information Engineering,Southwest University of Science , Technology,Mianyang,Sichuan ,China .National Key Laboratory of Communication,University of Electronic Science , Technology of China,Chengdu ,China .College of Electronics Engineering,University of Electronics Science , Technology of China,China.New method for edge detection of noisy gray scale image[J].Computer Engineering and Applications,2010,46(11):183-185. | |
| Authors: | CHEN Hong-yan MA Shang WANG Hai-jiangCollege of Information Engineering Southwest University of Science Technology Mianyang Sichuan China National Key Laboratory of Communication University of Electronic Science Technology of China Chengdu China College of Electronics Engineering University of Electronics Science Technology of China China |
| Affiliation: | CHEN Hong-yan1,MA Shang2,WANG Hai-jiang31.College of Information Engineering,Southwest University of Science , Technology,Mianyang,Sichuan 621010,China 2.National Key Laboratory of Communication,University of Electronic Science , Technology of China,Chengdu 610054,China 3.College of Electronics Engineering,University of Electronics Science , Technology of China,China |
| Abstract: | The normal methods with different algorithm are very sensitive to noise in edge detection.According to this problem,a new edge detection method for noisy gray scale image based on Independent Component Analysis (ICA) is proposed.The methods extract pattern template by calculating the higher statistical information of image data.Then a gray scale image contaminated by Gaussian noise is transformed into a pattern map by pattern matching.Thus the edges can be easily detected from pattern map.The experiments sh... |
| Keywords: | edge detection noisy image Independent Component Analysis(ICA) pattern template |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ | |
| 点击此处可从《计算机工程与应用》浏览原始摘要信息 | |
| 点击此处可从《计算机工程与应用》下载全文 | |
