Title :
A two-dimensional smoothing template for digital image preprocessing
Author :
Zhu, Hai-An ; Zheng, Nan-Ning ; Zhang, Xi-Ning
Author_Institution :
Artificial Intelligence & Robot Inst., Xi´´an Jiaotong Univ., China
Abstract :
The theories of least-squares estimation and nonlinear function fitting are applied to solve the problem of data smoothing in image processing. Based on the derived algorithm, a novel spatial-domain template with variable mask size is obtained. It can preserve the important information in digital images as it smooths. Experimental results and detailed case studies have shown that the derived template has advantages over the commonly adopted filtering masks. The quality of the resulting images is very good, even under strongly noise-corrupted conditions
Keywords :
computerised picture processing; least squares approximations; 2D smoothing template; digital image preprocessing; least-squares estimation; mask size; nonlinear function fitting; spatial-domain template; strongly noise-corrupted conditions; Digital images; Digital signal processing; Filtering algorithms; Image processing; Least squares approximation; Limiting; Noise reduction; Signal processing; Signal processing algorithms; Smoothing methods;
Conference_Titel :
Industrial Electronics Society, 1990. IECON '90., 16th Annual Conference of IEEE
Conference_Location :
Pacific Grove, CA
Print_ISBN :
0-87942-600-4
DOI :
10.1109/IECON.1990.149146
