基于DTCW T方向信息的超声图像斑点噪声消除  

Speckle Reduction for Ultrasound Image Using Edge Directions in DTCWT Domain

作　　者：许慰玲[1] 沈民奋[2] 杨金耀 

机构地区：[1]仲恺农业工程学院信息学院,广州510225 [2]汕头大学广东省数字信号与图像处理技术重点实验室,汕头515063 [3]汕头市超声仪器研究所有限公司,汕头515041

出　　处：《数据采集与处理》2011年第6期671-675,共5页Journal of Data Acquisition and Processing

基　　金：国家自然科学基金(61072037)资助项目;广东省科技厅科技计划(2010B060900111)资助项目

摘　　要：针对一般的小波去噪方法在去除超声图像斑点噪声时不能有效保持图像边缘信息的问题,本文提出基于双树复小波变换(Dual tree complex wavelet transform,DTCWT)方向信息的超声图像斑点噪声消除算法。利用双树复小波变换6个方向复小波系数的相对方差来确定该点是否位于边缘或纹理上,对低于门限的像素高频复系数置零以实现超声图像DTCWT域的自适应滤波。实验结果表明,与Lee,Frost,Kuan滤波方法相比,本文算法具有明显的优势;在图像平滑效果接近的情况下,本文算法边缘保持度略优于Yu提出的各向异性斑点噪声抑制(Speckle reducing anisotropic diffusion,SRAD)方法。In general, the edge information of an image can not be effectively preserved when traditional wavelet de-noising methods are adopted to reduce the speckle in ultrasound images. In view of this problem, a novel speckle reducing algorithm for ultrasound images is proposed based on the dual tree complex wavelet transform （DD_DTCWT） direction information. The relative variance of the six directional complex detail coefficients about DTCWT is used to de- termine whether points are located at the edges or in the textures. Specifically, speckle is re- duced by resetting complex detail coefficients to zero for the pixels beyond the threshold. Through such a way the adaptive filter in DTCWT domain is realized. Experimental results show that the proposed method significantly outperforms the state-of-the-art competitors in- cluding Lee, Frost and Kuan filtering methods. Also, the degree of edge maintaining of the method is better than the speckle reducing anisotropic diffusion （SRAD） method proposed by Yu for images with similar smoothing effects.

关 键 词：超声图像 斑点噪声 双树复小波变换 边缘方向 自适应滤波 

分 类 号：TP751[自动化与计算机技术—检测技术与自动化装置]