Multiview Image Block Compressive Sensing with Joint Multiphase Decoding for Visual Sensor Network

Abstract

In this article, a multiview image compression framework, which involves the use of Block-based Compressive Sensing (BCS) and Joint Multiphase Decoding (JMD), is proposed for a Visual Sensor Network (VSN). In the proposed framework, one of the sensor nodes is configured to serve as the reference node, the others as nonreference nodes. The images are encoded independently using the BCS to produce two observed measurements that are transmitted to the host workstation. In this case, the nonreference nodes always encoded the images (I NR ) at a lower subrate when compared with the images from the reference nodes (I R ). The idea is to improve the reconstruction of I NR using I R . After the two observed measurements are received by the host workstation, they are first decoded independently, then image registration is applied to align I R onto the same plane of I NR . The aligned I R is then fused with I NR , using wavelets to produce the projected image I P . Subsequently, the difference between the measurements of the I P and I NR is calculated. The difference is then decoded and added to I P to produce the final reconstructed I NR . The simulation results show that the proposed framework is able to improve the quality of I NR on average by 2dB to 3dB at lower subrates when compared with other Compressive Sensing (CS)--based multiview image compression frameworks.

Block Compressive Sensing (BCS) Based Low Complexity, Energy Efficient Visual Sensor Platform with Joint Multi-Phase Decoder (JMD)

“…However, in our research, we have used the joint multiphase decoding (JMD) framework for the reconstruction of images that make use of the TV minimization approach and is referred to as BCS-JMD-TV. A brief overview of the scheme is described as follows, the details of the scheme can be referred from [17,18]…”

“…BCS-JMD-TV [17,18] is a multi-view compression scheme for VSN based on block-based compressive sensing (BCS) and joint multi-phase decoding (JMD). First, images captured by different visual nodes are encoded using BCS to reduce the hardware complexity.…”

“…At the server, the received data is first decoded by BCS to recover the images. The images then go through the joint multi-phase decoding (JMD) [17] framework, where the correlations among the images are exploited. The exploited information is used to improve the visual quality of the decoded images.…”

“…In addition, the developed visual node is also capable of capturing images and recording videos. However, this research work is limited to multi-view image scenario, and BCS along with JMD framework [17,18,19,20] is implemented on low power applications (VSN) to evaluate and analyze its performance based on the reconstruction quality and computational complexity of BCS rather than measuring the bandwidth utilization.…”

Relative Pose Based Redundancy Removal: Collaborative RGB-D Data Transmission in Mobile Visual Sensor Networks

“…The spatial correlation is explored and removed at encoders by image registration algorithms. Only the uncorrelated visual content is delivered in the network after being jointly encoded by some recent coding techniques (e.g., Multiview Video Coding (MVC) [ 38 , 39 ]) and compressive sensing approaches [ 40 , 41 ]. However, at least one node in the network is required to have the full set of images captured by the other sensors in order to perform image registration.…”