Oguz Ipek, Malone Joseph D, Atay Yigit, Tao Yuankai K
Vanderbilt University, Nashville, TN.
Proc SPIE Int Soc Opt Eng. 2020 Feb;11313. doi: 10.1117/12.2549472. Epub 2020 Mar 10.
Reducing speckle noise is an important task for improving visual and automated assessment of retinal OCT images. Traditional image/signal processing methods only offer moderate speckle reduction; deep learning methods can be more effective but require substantial training data, which may not be readily available. We present a novel self-fusion method that offers effective speckle reduction comparable to deep learning methods, but without any external training data. We present qualitative and quantitative results in a variety of datasets from fovea and optic nerve head regions, with varying SNR values for input images.
减少斑点噪声是改善视网膜光学相干断层扫描（OCT）图像视觉和自动评估的一项重要任务。传统的图像/信号处理方法只能实现适度的斑点减少；深度学习方法可能更有效，但需要大量的训练数据，而这些数据可能无法轻易获得。我们提出了一种新颖的自融合方法，该方法能实现与深度学习方法相当的有效斑点减少，但无需任何外部训练数据。我们在来自中央凹和视神经乳头区域的各种数据集中呈现了定性和定量结果，输入图像的信噪比各不相同。
