摘要
为适应不同设备、不同信道和不同应用的要求,数字图像处理中需要改变图像的空间分辨率。传统的改变图像空间分辨率的办法是抽取和线性插值。本文基于小波分析提出由高分辨率图像可得到各种较低分辨率图像的多分辨方法符合人的视觉系统特性。为了提高图像的空间分辨率,本文根据小波子带间的相关性等特点,提出由低分辨率的高频细节预测高分辨率的高频细节方法,从理论分析到实验结果,均表明本文方法是可行的,而且优于传统的方法。
It is usually necessary to change the image resolution for different purpose in digital image processing. The traditional method is decimation and linear interpolation. The multiresolution image representation based on the wavelet analysis is introduced. By this way, the image resolution can be decreased in keeping with the human visual system characteristic. Based on the relativity among the wavelet subbands, the details of the higher resolution image can be predicted according to the details of the lower resolution image. Higher resolution image is then obtained by the wavelet synthesis. Experiment results show that the scheme is feasible and better than conventional algorithm.
出处
《江苏科技大学学报（自然科学版）》
CAS
北大核心
2005年第6期50-53,共4页
Journal of Jiangsu University of Science and Technology:Natural Science Edition
基金
江苏科技大学资助课题(校编2005DX006J)
关键词
数字图像
图像分辨率
小波变换
digital image
image resolution
wavelet transform
作者简介
李永忠（1961-），男，甘肃兰州人，江苏科技大学教授。
