Abstract—In this paper Short Time Fourier Transform (STFT) is used to despeckle the medical ultrasound image before reconstruction. STFT is applied individually to the one dimensional (1D) Radio Frequency (RF) envelopes that constitute the two dimensional (2D) image. Total Variation Filter (TVF) and Anisotropic Diffusion Filter (ADF) are also applied to the speckle noisy 2D ultrasound image after reconstruction. Performance comparison is held between despeckling the ultrasound image before reconstruction and after reconstruction. Despeckling the ultrasound image before reconstruction through STFT has removed the speckle noise more efficiently than TVF and ADF. It also maintained the texture of the original image and that is compared to STFT performance when applied to the 2D image after reconstruction. TVF is found more efficient in removing speckle when applied through overlapping blocks compared with applying it as a whole to the 2D image. ADF is found outperforming TVF in removing speckle noise and it is found more efficient when applied as a whole to the 2D image compared with applying it through overlapping blocks.

Index Terms—ultrasound image, RF envelope, down-sampling, STFT, speckle noise

Cite: Jawad F. Al-Asad, "Despeckling the 2D Medical Ultrasound Image through Individual Despeckling of the Envelopes of Its 1D Radio Frequency Echo Lines by STFT," Journal of Image and Graphics, Vol. 4, No. 2, pp. 67-72, December 2016. doi: 10.18178/joig.4.2.67-72