---

USING FCM-BASED PRE-CLASSIFICATION AND RIBM-OPTIMIZED NLM FILTER FOR ULTRASOUND IMAGE DESPECKLING

Abstract Ultrasound imaging is widely used in medical diagnostics due to its non-invasive nature and real-time capabilities. However, the presence of speckle noise significantly degrades image quality, making the accurate interpretation of anatomical structures challenging. Traditional despeckling methods often compromise edge preservation and fail to adapt to varying noise levels across different image regions. This study introduces a novel approach that integrates Fuzzy C-Means (FCM) clustering-based pre-classification with a Robust Intensity-Based Metric (RIBM)-enhanced Non-Local Means (NLM) filter to address these challenges. Initially, the FCM clustering algorithm pre-classifies the ultrasound image into distinct homogeneous and heterogeneous regions, enabling region-specific processing. The RIBM-enhanced NLM filter is then applied to each region, ensuring effective noise suppression while preserving critical image details. Experimental evaluation was conducted on a dataset comprising 50 clinical ultrasound images. Quantitative metrics such as Peak Signal-to-Noise Ratio (PSNR), Structural Similarity Index Measure (SSIM), and Edge Preservation Index (EPI) were used for performance assessment. Results demonstrate that the proposed method achieves superior despeckling performance, with an average PSNR of 34.12 dB, SSIM of 0.926, and EPI of 0.874, outperforming traditional NLM and wavelet-based methods. These results validate the efficacy of the proposed framework in enhancing ultrasound image quality while maintaining structural integrity. Authors N.P. Ponnuviji R.M.K. College of Engineering and Technology, India Keywords Ultrasound Despeckling, Fuzzy C-Means Clustering, Non-Local Means Filter, Robust Intensity-Based Metric, Medical Image Processing Yearly Full Views January February March April May June July August September October November December 0 8 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0

Published By : ICTACT Published In : ICTACT Journal on Image and Video Processing ( Volume: 15 , Issue: 3 , Pages: 3517 - 3522 ) Date of Publication : February 2025 DOI : 10.21917/ijivp.2025.0498 Page Views : 5 Full Text Views : 1