An adaptive filter for speckle reduction in medical ultrasound image processing

• Published in: ICARCV 2004 : 8th Control, Automation, Robotics and Vision Conference, 2004 : 6-9 December 2004 Vol. 1; pp. 654 - 658 Vol. 1
• Main Authors: Zhang, L.C., Wong, E.M.C., Koh, L.M., Ng, L.S.
• Format: Conference Proceeding
• Language: English
• Published: IEEE 2004
• Subjects: Adaptive filters; Biomedical engineering; Biomedical imaging; Image processing; Image resolution; Medical diagnostic imaging; Speckle; Statistics; Synthetic aperture radar; Ultrasonic imaging
• ISBN: 9780780386532; 0780386531
• DOI: 10.1109/ICARCV.2004.1468904

Speckle has been a major interference in medical ultrasound imaging. In this paper, an adaptive filter for speckle reduction is proposed. The new algorithm is based on the two-dimensional adaptive least-mean-square filter (TDLMS) design. In the new method, a simple weighted local dynamics is defined to verify the local speckle formation extent. The estimated error between the weighted local dynamics and the ideal local dynamics is used to control the filter's performance. Thus the filter would work between the maximum speckle smoothing and fine details preservation, by outputting the local average and the original value in the image, separately. The simple filter structure makes the new algorithm easy to implement. Simulation results show that the new algorithm performs well in both speckle reduction and details preservation.