UAV REAL-TIME DATA PROCESSING THROUGH FLIGHT TRAJECTORY AND RESAMPLE RATE ANALYSIS

In the recent years, the Unmanned Aerial Vehicles (UAVs) have become increasing interests for different civilian applications. The high-resolution video image, on one hand, brings us clarity and details of the behavior and characteristics of the earth surface features, on the other hand, presents new challenges in data processing. For example, how do we generate ortho-image from high-resolution UAV video images because the video data contain large redundant information due to a high sampling rate and the UAV has inherent characteristics of inconsistent and unstable flight parameters such as flying velocity, attitudes and altitude. This paper presents our theoretical analysis and experimental results for optimizing the re-sampling rate in order to speed up the generation of the video ortho-image. This algorithm is based on analysis of UAV exterior orientation parameters (EOPs) in each imaging epoch and video lens distortion characteristics as well as mapping accuracy of different scale. The model has been established for these relationships based on a test field located in Picayune, Mississippi. Our experimental results indicate that the resample rate is directly correlated with EOPs including three position elements and rotation elements of exterior orientation parameters, flying velocity and requirement of accuracy for different mapping scale.