Target Localization Accuracy Improvement in Wireless Visual Sensor Networks

Target localization that is to be aware of position of the nodes in a network is an essential issue in wireless visual sensor networks (WVSN). In these networks, each node using a low-resolution camera and some image processing algorithms, extracts multiple feature points to represent the target position at the sensor node level. In this paper, a statistical method of merging the position information of different sensor nodes to select the most correlated feature point pair at the base station is presented. This method releases the influence of the accuracy of target extraction on the accuracy of target localization in universal coordinate system. Then the results are improved based on including the cameras field of views (FOVs) common area. Simulation results confirm that the proposed method can generate more desirable target localization’s accuracy in comparison with the other target localization methods

کلیدواژه ها:

WSN ، WVSN ، WCSN ، CWN