Camera Selection for Reduction of Altitude Estimation Error Caused by Pixel Quantization

With recent advances in camera sensor technology and distributed processing, visual sensor networks are being utilized in many applications such as automatic surveillance, environmental monitoring and etc. Determining altitude of a target in a specific region is one of these applications. Due to quantization in CCD of camera nodes, the obtained information from these nodes is not very accurate. This error is due to many parameters in the network and can be decreased by selecting cameras at suitable positions. Existing methods improve this estimation but without considering the lifetime of the network. In this paper, we utilize a method that calculates the height-estimation-error of the target. Furthermore we propose an algorithm that selects a pair of cameras which have maximum remaining energy while the produced error is less than a defined threshold. It is shown through simulation that the performance of the proposed selection algorithm is compatible with that of the full search algorithm. Also, our algorithm has low computation overhead and it increases the lifetime of the network.