Despeckling of OCT images using DT-CWT based fusion technique

Abstract

Optical Coherence Tomography (OCT) is a non-invasive diagnostic method used to identify and manage a variety of retinal illnesses. However, the OCT image formation method gives rise to speckle noise that can drastically degrade the quality of OCT images. These low-quality images have a detrimental effect on subsequent illness diagnoses. There are numerous software-based speckle reduction techniques, each with its own advantages. Hence, combining even some of them maximizes the benefits. We propose a fusion approach based on spatial domain filtering and a sparsity-based denoising method for speckle noise removal in OCT images. Our hybrid methodology employs the Dual-Tree Complex Wavelet Transform (DT-CWT) to integrate both spatial domain filtering and the sparsity-based denoising method and uses a multi-scale fusion technique to merge the two transformed output images. Experimental findings indicate that the proposed algorithm could successfully combine two images, resulting in an output OCT image with significant speckle reduction and image structure preservation.