Analog Image Denoising with an Adaptive Memristive Crossbar Network

TL;DR

This paper introduces an adaptive analog in-memory neural network for image denoising that learns new noise patterns and can be integrated with CMOS sensors, demonstrating significant improvements in image quality metrics.

Contribution

It presents a novel adaptive denoising system using an analog memristive crossbar network with three configurations, enhancing noise removal and integration capabilities.

Findings

Single layer network: 3.2us processing time, 21nJ energy, 0.3mm^2 area

Convolution network: 72ms processing time, 236uJ energy, 0.48mm^2 area

Maximum improvements in SSIM, MSE, PSNR by 3.61, 21.7, and 7.7 times respectively

Abstract

Noise in image sensors led to the development of a whole range of denoising filters. A noisy image can become hard to recognize and often require several types of post-processing compensation circuits. This paper proposes an adaptive denoising system implemented using an analog in-memory neural computing network. The proposed method can learn new noises and can be integrated into or alone with CMOS image sensors. Three denoising network configurations are implemented namely, (1) single layer network, (2) convolution network, and (3) fusion network. The single layer network shows the processing time, energy consumption, and on-chip area of 3.2us, 21nJ per image, and 0.3mm^2 respectively, meanwhile, the convolution denoising network correspondingly shows 72ms, 236uJ, and 0.48mm^2. Among all the implemented networks, it is observed that performance metrics SSIM, MSE, and PSNR show a maximum improvement of 3.61, 21.7, and 7.7 times respectively.