Metadata-Assisted Global Motion Estimation for Medium-Altitude Unmanned Aerial Vehicle Video Applications

Abstract

Global motion estimation (GME) is a key technology in unmanned aerial vehicle remote sensing (UAVRS). However, when a UAV's motion and behavior change significantly or the image information is not rich, traditional image-based methods for GME often perform poorly. Introducing bottom metadata can improve precision in a large-scale motion condition and reduce the dependence on unreliable image information. GME is divided into coarse and residual GME through coordinate transformation and based on the study hypotheses. In coarse GME, an auxiliary image is built to convert image matching from a wide baseline condition to a narrow baseline one. In residual GME, a novel information and contrast feature detection algorithm is proposed for big-block matching to maximize the use of reliable image information and ensure that the contents of interest are well estimated. Additionally, an image motion monitor is designed to select the appropriate processing strategy by monitoring the motion scales of translation, rotation, and zoom. A medium-altitude UAV is employed to collect three types of large-scale motion datasets. Peak signal to noise ratio (PSNR) and motion scale are computed. This study's result is encouraging and applicable to other medium-or high-altitude UAVs with a similar system structure.

Geo-Location Algorithm for Building Targets in Oblique Remote Sensing Images Based on Deep Learning and Height Estimation

“…The error between the true position of the target (T R E = x r y r z r ) and the positioning result (T D E = x d y d z d ) can be calculated using the spatial two-point distance formula, as shown in Equation (23). 2 (23)…”

“…To realize remote sensing photogrammetry, various photoelectric sensors have been carried out on aircrafts to obtain ground images. Obtaining the location information of the target in the image using a geo-location algorithm is a research hotspot in recent years [1][2][3]. Currently, research on target positioning algorithms focuses on improving the positioning accuracy of ground targets, implying that the positioning error caused by the building height is rarely considered in the algorithm.…”

Adaptive Equalizer Design for Unmanned Aircraft Vehicle Image Transmission over Relay Channels

“…The experimental UAV images, with the size of 1392 × 1040 pixels [22], are captured by a mediumaltitude UAV that can cruise at an altitude of 1000 m. The M-UAV plans to transmit the image to GS via R-UAV. The distance between M-UAV and R-UAV and that between R-UAV and GS are both assumed to be 5 km.…”

Sensor-Assisted Global Motion Estimation for Efficient UAV Video Coding

“…The motion of each coding unit in UAV video stream is composed of camera and foreground object motion. GME is a technique that attempts to find the perceptive projection matrix between two images for video processing of high-mobility systems [2]. Differing from region ME which attempts to find the corresponding position of each individual pixel in its reference pictures, GME identifies the background motion introduced by the camera to obtain a stable and smooth video.…”

“…The image frames might move out of the search window or image distortion caused by rotation/zoom might result in prediction unit block matching algorithms failure. The compression ratio deteriorates significantly if the block matching algorithm fails.To provide a global motion model that fits a wide range of mid-altitude UAVs, the authors of [2] propose to derived image coordinate system transform model from metadata. A low delay and low complexity video encoding for UAV inspection application is presented in [8] which replace inter-frames using homography matrix.…”