Area coverage is an important research issue in the field of visual sensor networks (VSNs) because of the inherent constraints of VSNs, such as non-rechargeable energy resources and directionality of the sensing range of camera nodes. The dense deployment of camera nodes makes it possible to provide a satisfactory area coverage for a longer duration. At the same time the rest of camera nodes can be turned off and be scheduled to alternate the active nodes when it is necessary. In this paper, we define area coverage problem in VSNs aiming to minimize blind and redundantly covered grid cells of a desired area and energy distortion of camera nodes. Then we propose two scheduling algorithms for camera nodes which are randomly deployed to k-cover the desired area. In the first algorithm named evolutionary camera node scheduling (ECNS), we aim to achieve maximal area coverage by putting the smallest number of camera nodes into active mode and to minimize blind and redundantly grid cells. Since the objectives considered in ECNS conflict each other, we employ adaptive weighted sum method to formulate our objectives into a linear equation and then we propose a genetic algorithm to find the minimum value of the integrated linear equation. In the second algorithm named energy aware evolutionary camera node scheduling (EAECNS), we propose a method to strike a balance between the energy consumption of all camera nodes while it is providing satisfactory coverage of the target area and keeping the number of redundantly covered grid cells down. We evaluate the performance of both algorithms in terms of coverage, number of live nodes and redundancy by subsequent simulations. Also, we show that EAECNS has superior performance in comparison with ECNS and other state-of-the-art algorithms.
