一种Sobel算子的抗噪型边缘检测算法  被引量:2

An suppressing noise edge detection algorithm based on Sobel

在线阅读下载全文

作  者:沈德海[1] 张龙昌[1] 鄂旭[1] 

机构地区:[1]渤海大学信息科学与技术学院,辽宁锦州121013

出  处:《信息技术》2015年第5期81-84,80,共5页Information Technology

基  金:辽宁省高等学校实验室项目(L2012397);博士后基金项目(2012M520158);辽宁省"百千万人才工程"资助项目(2012921058);教育厅科研一般项目(L2012400)

摘  要:边缘检测在图像处理过程中占有重要的地位,Sobel算子是在数字图像边缘检测中常用的一种方法。经典Sobel算法简单、速度快,但也存在着边缘定位不精确、提取的边缘较粗、噪声干扰情况下抑制能力差等问题,针对这些问题,提出了一种抗噪声的Sobel边缘检测算法。算法先对图像采用多子窗口进行滤波,去除图像中存在的噪声;然后采用改进的Sobel算法对图像进行边缘检测,算法结合边缘方向计算梯度图像,并对梯度图像在3×3邻域内采用统计信息结合梯度阈值进行了2次边缘细化处理。与经典的Sobel算法及其他文献算法进行了对比试验,结果表明,该算法对噪声具有较强的抑制能力,在去除噪声的同时能够准确地检测出图像的边缘,而且得到的边缘更细,定位更精确。Edge detection is very important in the process of image processing, Sobel operator is a method that is commonly used in digital image edge detection. The classical Sobel algorithm is simple, fast, but also exists edge position is not accurate, extract the edge of the coarser and lower retrain ability for noise. To solve these problems, an suppressing noise algorithm based on Sobel is proposed. It removes the noise in the image with the multiple sub-windows filter algorithm firstly, then detects the edge by improved Sobel algorithm, the algorithm combining with the image edge direction, and the gradient image is refined twice within 3 × 3 neighborhood using statistical information combined with the gradient threshold. The results show that the proposed algorithm has stronger constrain ability for noise, detects the image edge accuracy, and obtains the finer edge than Sobel algorithm and others, the position is more accurate.

关 键 词:边缘检测 SOBEL算子 噪声 统计信息 边缘细化 

分 类 号:TP391.41[自动化与计算机技术—计算机应用技术]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关的主题
相关的作者对象
相关的机构对象