Pricise Target Geolocation and Tracking Based on Uav Video Imagery

Abstract

ABSTRACT:There is an increasingly large number of applications for Unmanned Aerial Vehicles (UAVs) from monitoring, mapping and target geolocation. However, most of commercial UAVs are equipped with low-cost navigation sensors such as C/A code GPS and a lowcost IMU on board, allowing a positioning accuracy of 5 to 10 meters. This low accuracy cannot be used in applications that require high precision data on cm-level. This paper presents a precise process for geolocation of ground targets based on thermal video imagery acquired by small UAV equipped with RTK GPS. The geolocation data is filtered using an extended Kalman filter, which provides a smoothed estimate of target location and target velocity. The accurate geo-locating of targets during image acquisition is conducted via traditional photogrammetric bundle adjustment equations using accurate exterior parameters achieved by on board IMU and RTK GPS sensors, Kalman filtering and interior orientation parameters of thermal camera from pre-flight laboratory calibration process. The results of this study compared with code-based ordinary GPS, indicate that RTK observation with proposed method shows more than 10 times improvement of accuracy in target geolocation.

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

“…The X-axis of the aircraft frame points to the nose of the aircraft, the Y-axis points to the right wing, and the Z axis points downward to form an orthogonal right-handed set. The attitude of the camera can be described by the inner and outer frame angles, pitch θ and roll θ , respectively, and the transformation matrix of the camera frame and aircraft frame is as expressed in Equation (14).…”

“…Hosseinpoor improves the accuracy of target positioning through multiple measurements. This filtering-based positioning algorithm is widely used in target tracking [14].…”

Detection, Tracking, and Geolocation of Moving Vehicle From UAV Using Monocular Camera

“…Regarding the generative model methods, such as the particle filter [15] or mean-shift tracking [16], the target region is modeled in the current frame, and then, the most similar region to the model will be predicted as the target pixels' position. For example, Tehran University proposed a visual tracking and geolocation project and chose the mean-shift as the visual tracking solution [6]. These methods are not appropriate for the video captured from a UAV using a visible light camera because they are susceptible to the complex background; they are only suitable for the cases where the pixel size of the object is relatively large and the object moves at a low speed.…”

“…In another tracking and geolocation system [6], the realtime kinematic (RTK) global positioning systems (GPS) and thermal camera were installed on the multirotor UAV. The stationary object was chosen manually, and the object visual tracking was implemented through the mean-shift method.…”

“…The stationary object was chosen manually, and the object visual tracking was implemented through the mean-shift method. However, the estimation of the moving object was not presented in [6]. Moreover, the ground reference points must be set in advance for geolocation.…”