Abstract: The inherent speckle noise in ultrasound images severely degrades the image interpretation and affects the diagnostic results because of the characteristics of coherence in ultrasonic imaging.The speckle noise also increases the difficult for the following：up image processing tasks，such as edge detection，feature extraction and so on.Thus，it’s very important that reserves the edge characters effectively based on speckle noise removing.This paper divides the ultrasound image into uniform，speckle including and edge regions by studying the characters of speckle index C in neighborhood，then handles these regions differently.The simulation results prove that this methord is simple and valid，it removes the speckle noise and reserves the edge characters at the same time.The peformance of the method is better than the traditional median filtering algorithm.

Key words: ultrasound image denoising, speckle index C, median filtering algorithm

摘要： 超声成像的相干特性导致了超声图像中固有的斑点噪声，斑点噪声既降低了图像的质量影响诊断结果，又给图像的边缘检测、特征提取等后续处理带来困难。因此研究如何在去除斑点噪声的基础上，有效保留边缘特征具有十分重要的意义。通过研究邻域斑点指数C的特点，将超声图像划分为均匀区域、含斑区域以及边缘区域，对于不同区域采用不同的处理方法。仿真实验表明，该算法简单有效，在去除斑点噪声的同时更好地保留了边缘特征，其性能大大优于传统的中值滤波算法。

关键词: 超声图像去噪, 斑点指数C, 中值滤波算法