Optical coherence tomography (OCT) has shown great promise for non-invasive and contact-less imaging of subsurface soft tissues. However, the problem of low image contrast caused by speckle noise has limited its applications in diagnosis. This paper present a wavelet-based method designed to improve OCT image contrast by reducing speckle noise. After transforming the image into a set of sub-images with different resolution levels in wavelet domain, we threshold high-frequency coefficients nonlinearity in horizontal, vertical, and diagonal directions. The experimental results show that wavelet processing suppresses speckle noise in OCT images of soft tissue effectively, while maintaining the sharpness of image features.© (1998) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.