Monitoring or surveillance needsUnmanned Aerial VehiclesorDronesto cover each point within the area of interest. In general, any outdoor region can be modeled as a surface. Thus we focus on surveillance of a surface with a random connected distribution of drones such that a reorganization of drones maximizes the covered area without creating any coverage hole. In addition, the target is to provide a compact coverage to minimize the diameter of the network formed by the drones, which helps in faster communications. We present a centralized algorithm that achieves amaximal compactcoverage. We propose two distributed algorithms based onVirtual ForceandLocal Voronoi, respectively. Extensive simulation studies show that our proposed algorithms result in hole-freemaximal compactcoverage with limited displacement. The self-organizing behavior of ourLocal Voronoibased algorithm can be replicated even in the presence of failures in the drones. The algorithm is capable of recovering a maximal coverage with the remaining non-faulty drones. Also, it is robust with respect to the addition of new drones in real-time. We finally supplement the simulation results to include churn in the network in the context of fault-tolerance and scalability.