MODE FILTERING TO REDUCE ULTRASOUND SPECKLE FOR FEATURE-EXTRACTION

The authors investigate the use of filtering techniques to reduce spec kle in ultrasound images, to improve their suitability for later featu re extraction. The maximum likelihood estimator for a speckle corrupte d image is shown to correspond to the statistical mode but this is dif ficult to determine for small populations, such as those contained by a filter mask. The truncated median filter approximates the mode by us ing the order of known image statistics and provides a fully automated image processing technique for speckle filtering. The filter's perfor mance is established using a new quantitative evaluation scheme that c losely considers the effect of filtering on edges, a key factor when a pplying features extraction in automated image interpretation. Applica tion to in vivo and phantom test images shows that the truncated media n filter provides clear images with strong edges, of quality exceeding that of other techniques. These benefits are confirmed by the applica tion of feature extraction in arterial wall labelling.