基于形态学多结构基元的含噪图像边缘检测  被引量:8

Noise Image Edge Detection Based on Morphological Multi-structural Elements

在线阅读下载全文

作  者:夏平[1,2] 刘馨琼[1,2] 向学军[2] 万钧力[2] 

机构地区:[1]三峡大学智能视觉与图像信息研究所,湖北宜昌443002 [2]三峡大学电气信息学院,湖北宜昌443002

出  处:《计算机仿真》2010年第7期206-209,共4页Computer Simulation

基  金:湖北省教育厅自然科研项目(D200513001);宜昌市科技发展计划项目(A2007107-08)

摘  要:为了研究基于多结构基元构成形态学结构基,提出了一种图像边缘检测算法。首先,定义由开-闭运算、以及闭-开运算加权组合形成复合形态学滤波器,应用该滤波器对图像进行滤波;其次,不同形状的结构元素能检测出不同方向和结构的边缘信息,对定义多方向结构元素以组成形态学结构基,应用此结构基对滤波后的图像进行边缘检测。通过仿真重建形成理想的图像边缘,仿真结果表明,应用于含噪图像边缘检测算法,使抗噪的MSE性能较"开-闭运算"方法减少了7.82%和6.38%,PSNR性能提高了14.68%和8.05%,在检测精度方面得到了连续和封闭的边缘信息,与经典算子检测算法比较边缘信息更清晰。An image edge detection algorithm based on multi - structural morphological elements is proposed.The algorithm firstly defines the closing - opening operators as well as the compound morphology filter with opening -closing operators,and applies the morphology filter to the image.Secondly,because the different shape structural element can detect different direction and the structure edge information,this algorithm defines the multi - direction structural element composition morphology structure base,and applies the structure base to detect image edge information to reconstruct ideal image edge.Experimental results show that the algorithm can be applied to detect the noise image edge,of which the MSE performances of image denoising are respectively less than the opening - closing operators by 7.82%and closing - opening operators by 6.38%,and the PSNR performances are respectively better than the opening - closing operators by 14.68%and closing - opening operators by 8.05%.In the aspect of detection precision,the continuous and closed edge information can be obtained.Comparing with the traditional operators detection algorithm,the new algorithm can get clearer edge.

关 键 词:边缘检测 数学形态学 多结构基元 算子 

分 类 号:TN911.73[电子电信—通信与信息系统] TP391.9[电子电信—信息与通信工程]

 

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