Skip to Main Content
In this paper, we address the problem of localization-oriented optimal cameras selection based upon a tradeoff between the accuracy of target localization and the energy consumption in camera sensor networks. We model the addressed problem into the combined optimization problem of selecting a sequence of sets to minimize the cost while achieving a specified utility. Using the vision-based localization model, we construct an entropy-based utility function. This utility function quantizes reduction in the entropy of the target location distribution according to the fusion of the selected camera observations with the prior target location distribution. Furthermore, we design an algorithm to find the optimal subset of cameras based on the criteria: minimizing the cost while attaining a specified accuracy of localization. We conduct experiments to validate and evaluate our proposed scheme.