NeighCNN: A CNN based SAR Speckle Reduction using Feature preserving Loss Function

TL;DR

NeighCNN is a deep learning-based SAR speckle reduction method that uses a novel combined loss function to effectively remove noise while preserving image edges and details, validated on synthetic and real SAR images.

Contribution

The paper introduces NeighCNN, a CNN architecture with a unique combined loss function for improved speckle noise reduction in SAR images.

Findings

Effective noise removal demonstrated on synthetic and real SAR images

Superior edge preservation compared to existing methods

Improved image quality metrics such as PSNR and SSIM

Abstract

Coherent imaging systems like synthetic aperture radar are susceptible to multiplicative noise that makes applications like automatic target recognition challenging. In this paper, NeighCNN, a deep learning-based speckle reduction algorithm that handles multiplicative noise with relatively simple convolutional neural network architecture, is proposed. We have designed a loss function which is an unique combination of weighted sum of Euclidean, neighbourhood, and perceptual loss for training the deep network. Euclidean and neighbourhood losses take pixel-level information into account, whereas perceptual loss considers high-level semantic features between two images. Various synthetic, as well as real SAR images, are used for testing the NeighCNN architecture, and the results verify the noise removal and edge preservation abilities of the proposed architecture. Performance metrics like peak-signal-to-noise ratio, structural similarity index, and universal image quality index are used for evaluating the efficiency of the proposed architecture on synthetic images.