An efficient transmitting strategy for image fusion in WMSN

Wireless Multimedia Sensors Networks are characterized by a high consumption of network resources such as energy and bandwidth. This consumption is caused by the large amount of data flowing across the network. One of the mechanisms used to reduce the consumption of resources is data fusion to eliminate redundant transmission while synthesizing the relevant information. However, as data fusion can operate only on specific nodes, called fusion agents, an important amount of data still remains sent from sensor nodes to the fusion agents. In this paper, we deal with the fusion of Multi-focus images. In such applications the totality of the images are sent from the cameras to the fusion agents before to be fused, which results in a huge consumption of the network resources. To tackle this issue, we propose a transmitting strategy that determines the useful information to send. This is achieved by computing different similarity coefficients at different levels. The values of the similarity thresholds are adaptable making it possible to cope with the context of the network and the running application. Primary simulation results show the effectiveness of the proposed mechanism in terms of network resources consumption and delay transmission.