An Improved Approach to Image De-Noising Based on Multi-wavelet and Threshold
Image de-noising is widely used in image pre-processing. Its aim is to enhance signal-to-noise ratio (SNR), improve the quality of the image and project the expected feature of the image. The better de-noising result depends on the degree of the noise and image details reserved. In fact, image de-noising is based on the fact that noise and energy of the image are distributed within different frequency band. Generally, its energy is distributed over low frequency band, while both its noise and details are distributed over high frequency band.Corresponding Semi-soft threshold is used in different scale high frequency sub-bands in this paper. In this paper, noised image file is decomposed by multi-wavelet transform first; and then, the coefficients are processed with different semi-soft thresholds according to coefficients energy distribution. The reconstructed image can be obtained by using the inverse multiwavelet transform. Simulation experiments show that image noise can be reduced effectively and image details can be preserved a lot by this method and it should be a better improved method.
S Q Zhang X H Xu J T Lv X Y Zang N He
Electrical Engineering Institute,Yanshan University, Qinhuangdao, China, 066004
国际会议
第四届仪器科学与技术国际会议( 4th International Symposium on Instrumentation and Science and Tcchnology)
哈尔滨
英文
412-416
2006-08-08（万方平台首次上网日期，不代表论文的发表时间）