Image sequences filtering using a new nonlinear method

被引：0

作者：

Saeidi, M ^{[1
]}

Motamedi, SA ^{[1
]}

Saeidi, B ^{[1
]}

Behrad, A ^{[1
]}

Saeidi, R ^{[1
]}

机构：

[1] Iran Telecommun Res Ctr, Polytech Elect Engn Dept, Tehran, Iran

来源：

PROCEEDINGS OF THE FIFTH IASTED INTERNATIONAL CONFERENCE ON VISUALIZATION, IMAGING, AND IMAGE PROCESSING | 2005年

关键词：

image sequences filtering; non-linear functions; low-density noise; median filter;

D O I：

暂无

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

In this paper, we will propose a novel spatiotemporal non-linear method in image sequences filtering. The Median filter, using consecutive frames is more efficient than using only one frame in removing medium and high-density Salt and Pepper noise. However, simple median filter using only one frame removes low-density Salt and Pepper noise more effectively than using consecutive frames. This results from the fact that, the median filter isn't able to efficiently utilize the correlation of pixels among the consecutive frames under low-density Salt and Pepper noise. We will propose a solution to improve the performance of median filters using image sequences under low-density Salt and Pepper noise. The proposed method utilizes non-linear functions. to improve the efficiency of median filters using consecutive frames. Our proposed filter applies consecutive median to the selected pixels in each spatiotemporal window and then applies consecutive non-linear functions.

引用

页码：349 / 354

页数：6