Improved noise removal in fetal images using filter based preprocessing operation

Abstract

Speckle is a major ultrasound noise, which plays a key role in the image quality degradation and makes it difficult to
determine the image quality. Filtering that removes noises and reproduces ultrasound images is essential for improving
quality. Also, it is very necessary to maintain a good balance between the preservation of the characteristics and the
suppression of spikes using the filters used. On the other hand, it also observes the stronger ability of filters to suppress
speckles, but also keeps edge details of fetal image. In this paper, an effective pre-processing filtering operation using
different filters namely Order Statistics Filter, Median Filter, Entropy Filter, Gaussian Filter and Guided Filter is used to
eliminate the presence of speckle noise. The performance evaluation is carried out using three different metrics namely
MSE, SNR and PSNR. The result shows that the Gaussian filtering attains better performance than other methods.