Highly Efficient Speckle Noise Removal in Medical Images Using GSO Optimization
S. L. Shabana Sulthana, Sucharitha · 2021
Biomedical imaging assumes a significant function in different applications, such as early diagnosis, surgery, and classification. In general, the captured image is affected by noise, voltage, temperature fluctuations, and environmental conditions. Before image processing, the image must be pre-processed and unwanted pixel deviation should be removed to obtain a more accurate result. In image classification, the accuracy is directly proportional to the quality of the input image. This paper suggests a new way to remove speckle noise from input images using Glowworm Swarm optimization (GSO) and Wavelet Packet Transformation (WPT). The proposed technique is tested with several sample images and results were obtained with improved quality metrics. Various quality metrics such as Mean Square Error (MSE), Structural Similarity Index Metrics (SSIM), and Mean Absolute Error (MAE) are used to evaluate performance. From the analysis of the results, the proposed technique accomplishes better visual quality than the traditional methods.