A New Multi-quality Image Fusion Method in Visual Sensor Network

Abstract

Bandwidth, energy consumption and delay are always bottlenecks of visual sensor network. In this paper a hierarchical multi-quality image fusion method based on region-mapping is proposed to improve above condition. Camera view is divided into regions according to mapping relation and the structured deployment of nodes. The methods of motion attention analysis and nearest neighbor sampling are also adopted to optimize the local attention region of the image. The proposed fusion method is analyzed and verified to be effective to improve the quality of the monitor image with no need for more energy and bandwidth.