An Adaptive Speckle Suppression and Edge Enhancement Technique

Guo Shengwen,Luo Limin
DOI: https://doi.org/10.1007/s11767-003-0045-3
2003-01-01
Abstract:Speckle degrades severely the quality of medical B-scan ultrasonic images, especially it blurs edges and details of images. An adaptive speckle suppression and edge enhancement method based on Nakagami distribution is presented. The statistics of log-compressed echo images is derived for Nakagami distribution. An adaptive filter based on local statistical property of speckle is designed. The stick technique that utilizes sticks with different sizes and various orientations is applied to locally approximate certain linear features of image. The local region is a stick instead of a usual window, the orientation of sticks is decided by hypothesis test optimizing method and the length of sticks is obtained by region growing technique. Performance of the new method has been tested on the phantom and ultrasound images of pig muscle and echocardiographic. The results show that the technique effectively reduces the speckle noise while preserving and enhancing the tissue edge and resolvable details.
What problem does this paper attempt to address?