期刊刊名：管理科學與統計決策 　 卷期：7卷1期

篇名出版日期：2010年3月1日

作者：Lingling Jiang

語言：English

關鍵字：wavelet transform,medical ultrasonic image,threshold de-noising,noise statistical nature

被點閱次數：0次

閱讀時間：0sec

摘要：
De-noising characteristics of soft threshold, semi-soft threshold, hard threshold and noise statistical nature in wavelet transform are analyzed; Based on this a new De-noising algorithm of medical ultrasonic image based on wavelet transform is proposed. Towards borderline blur of soft threshold algorithm, linear operation is carried out to wavelet coefficients according to maximum of wavelet coefficients in each sub-band,meanwhile, high frequency wavelet coefficients bulge of semi-soft threshold algorithm is improved. Experimental results show that the detail information is reserved well while speckle noises are efficiently removed. Also the whole visual effect is improved.

　　　　　[全文下載]