Image Compressed Sensing with Multi-scale Dilated Convolutional Neural Network

TL;DR

This paper introduces a multi-scale dilated convolutional neural network framework for image compressed sensing that avoids block effects and captures multi-scale features, leading to improved reconstruction quality.

Contribution

The proposed MsDCNN framework jointly trains measurement and reconstruction networks without block division and employs multi-scale feature extraction to enhance image reconstruction.

Findings

Outperforms state-of-the-art methods in PSNR and SSIM

Effectively avoids blocking artifacts in reconstruction

Enhances feature extraction with multi-scale architecture

Abstract

Deep Learning (DL) based Compressed Sensing (CS) has been applied for better performance of image reconstruction than traditional CS methods. However, most existing DL methods utilize the block-by-block measurement and each measurement block is restored separately, which introduces harmful blocking effects for reconstruction. Furthermore, the neuronal receptive fields of those methods are designed to be the same size in each layer, which can only collect single-scale spatial information and has a negative impact on the reconstruction process. This paper proposes a novel framework named Multi-scale Dilated Convolution Neural Network (MsDCNN) for CS measurement and reconstruction. During the measurement period, we directly obtain all measurements from a trained measurement network, which employs fully convolutional structures and is jointly trained with the reconstruction network from the input image. It needn't be cut into blocks, which effectively avoids the block effect. During the reconstruction period, we propose the Multi-scale Feature Extraction (MFE) architecture to imitate the human visual system to capture multi-scale features from the same feature map, which enhances the image feature extraction ability of the framework and improves the performance of image reconstruction. In the MFE, there are multiple parallel convolution channels to obtain multi-scale feature information. Then the multi-scale features information is fused and the original image is reconstructed with high quality. Our experimental results show that the proposed method performs favorably against the state-of-the-art methods in terms of PSNR and SSIM.