Unmanned aerial vehicles (UAVs) equipped with cameras are a valuable tool for surveillance, reconnaissance, and protection of civilians, soldiers, and real estates. Multicopters or fixed wing UAVs patrol while an automatic video change detection localizes relevant or suspicious changes in the scene between two patrols. In this way, e.g., a convoy can be protected from improvised explosive devices (IEDs) by early detecting deployment traces like excavations, skid marks, footprints, left-behind tooling equipment, and marker stones. Furthermore, in case of disasters imminent danger can be recognized quickly. Therefore, an appropriate video change detection algorithm was realized recently as a solution. Since then, two main improvements could be realized which are described in this paper. First, a novel measurement for image differences in color space is introduced that increases the detection sensitivity. Furthermore, a solution is presented to eliminate or reduce detections of cast shadows in situations where the sun intensity and/or position is slightly or strongly different in the two compared patrols. In order to do this, the impact of cast shadows is examined in Lab and LCh color space to build up a dedicated shadow model with which shadows can be filtered out. This shadow model covers the relation between image intensity reduction, color shift towards blue, and image noise influences of cast shadows. The given results document the performance of the presented approach in different situations.