Full-view coverage holes detection and healing solutions

In this paper, we address the full-view coverage holes detection problem and healing solutions. Among many problems confronted in designing camera sensor networks (CSNs), the coverage challenge stands out as one of most fundamental problems. Some CSNs applications such as environment monitoring and animal tracking require covering each point of the target area. However, even if we can deploy camera sensor nodes to make the entire target region fully full-view covered at the very beginning, the nodes can die due to battery drain or environmental causes, which may generate full-view coverage holes in the region. Also, nodes may deviate from their initially assigned locations due to uncontrollable elements, leaving some areas not full-view covered. Full-view coverage holes reduce the ability of CSNs to detect events and network reliability. Consequently, it is critical to equip sensor nodes with efficient hole detection and recovery capabilities to ensure whole full-view coverage of the target field. We develop a novel full-view coverage analytical theory to address the full-view coverage holes detections problem, where coordinate system is applied to manipulate equations for fullview covered points, curves, and area. On the other hand, fullview coverage holes can be detected by this theory. The hole's boundary can be represented precisely by coordinates and the area of holes can be calculated quantitatively. This theory can also be utilized to study the relationship between the sensing range or the effective angle of cameras and full-view coverage area, therefore, it can serve as a guideline for designing full-view coverage camera sensor networks.