DESPECKLING OF POLYCYSTIC OVARY ULTRASOUND IMAGES BY IMPROVED TOTAL VARIATION METHOD

摘要

Ultrasound imaging provides a non-invasive, low cost, real-time imaging that helps the clinician in diagnosis and planning of therapy. However, the usefulness of ultrasound imaging is degraded by the presence of signal depended noise known as speckle. Image denoising algorithms are challenging operation because the fine details are embedded in medical image and diagnostic information should not be destroyed during noise removal. This paper presents speckle reduction by an improved total variation filter (ITV) method. The performance of these filters is evaluated based on the parameters such as mean square error (MSE), peak signal-to?noise ratio (PSNR), similar structure index mean (SSIM) and feature structure index mean (FSIM). An experimental result shows that the speckle noise can be efficiently removed by the ITV method without affecting the structure of the object.