Occlusion caused by vegetation is an essential problem for remote sensing applications in areas, such as search and rescue, wildfire detection, wildlife observation, surveillance, border control, and others. Airborne Optical Sectioning (AOS) is an optical, wavelength-independent synthetic aperture imaging technique that supports computational occlusion removal in real-time. It can be applied with manned or unmanned aircrafts, such as drones. In this article, we demonstrate a relationship between forest density and field of view (FOV) of applied imaging systems. This finding was made with the help of a simulated procedural forest model which offers the consideration of more realistic occlusion properties than our previous statistical model. While AOS has been explored with automatic and autonomous research prototypes in the past, we present a free AOS integration for DJI systems. It enables bluelight organizations and others to use and explore AOS with compatible, manually operated, off-the-shelf drones. The (digitally cropped) default FOV for this implementation was chosen based on our new finding.
翻译:植被造成的封闭是搜索和救援、野火探测、野生生物观测、监视、边界控制等遥感应用方面的一个基本问题。空载光学部分是一种光学、波长独立合成孔径成像技术,有助于实时消除计算隔离,可适用于无人驾驶飞机或无人驾驶飞机,如无人驾驶飞机。在本篇文章中,我们展示了森林密度与应用成像系统视野(FOV)之间的关系。这一发现是在模拟程序森林模型的帮助下作出的,该模型提供了比我们以前的统计模型更现实的隔离特性的考虑。虽然过去用自动和自主研究原型对AOS进行了探索,但我们为DJI系统提供了一种免费的AOS整合。它使蓝光组织和其他人能够使用兼容的、手动操作的离体无人机来使用和探索AOS。(数字裁剪裁的)用于这一执行的默认FOV。根据我们的新发现选择了(数字裁剪裁的)默认FOV。