Data-Guided Aerial Tracking

2020 ◽  
Vol 43 (8) ◽  
pp. 1540-1549
Author(s):  
Soumya Vasisht ◽  
Mehran Mesbahi
Keyword(s):  
Author(s):  
Changhong Fu ◽  
Ziang Cao ◽  
Yiming Li ◽  
Junjie Ye ◽  
Chen Feng
Keyword(s):  

2006 ◽  
Vol 84 (8) ◽  
pp. 1096-1103 ◽  
Author(s):  
J. Mainguy ◽  
G. Gauthier ◽  
J.-F. Giroux ◽  
I. Duclos

Many precocial birds make long-distance movements with their young after hatch to reach the best foraging sites. On Bylot Island, Nunavut, a large number of Greater Snow Goose ( Chen caerulescens atlantica L., 1758) families move 30 km from the main nesting colony (MNC) to reach the main brood-rearing area (MBR) soon after hatch. Geese moving from the MNC to the MBR generally rear lighter and smaller goslings than geese that avoid this movement by both nesting and rearing their brood at the MBR. In this study, we tested the hypotheses that use of low-quality habitats and an increase in the time spent walking at the expense of foraging during movements could explain the reduced growth of goslings in those families. We conducted visual observations to compare habitat use and selection as well as behaviour of geese during brood movements from the MNC to the MBR (i.e., at a transit area) with those of families that had already settled at the MBR. We also conducted aerial tracking to monitor habitat use of 16 radio-marked females during and after brood movements. Streams, wet polygons, and lakes, considered high-quality habitats in terms of feeding opportunities and predator refuges, were preferred, while upland, a low-quality habitat, was avoided at both the transit area and the MBR. However, broods were found in the upland habitat more often during movements than once settled on a rearing site. The behaviour of unmarked geese at the transit site did not differ from that of geese at the MBR. We suggest that reduced food intake in low-quality habitats during movements, but not the increase in time spent walking, may explain the reduction in growth observed at fledging in goslings moving from the MNC to the MBR.


2016 ◽  
Vol 66 (2) ◽  
pp. 122 ◽  
Author(s):  
Vindhya P. Malagi ◽  
Ramesh Babu D.R. ◽  
Krishnan Rangarajan

<p>Vison based tracking in aerial images has its own significance in the areas of both civil and defense applications.  A novel algorithm called aerial tracking learning detection which works on the basis of the popular tracking learning detection algorithm to effectively track single and multiple objects in aerial images is proposed in this study. Tracking learning detection (TLD) considers both appearance and motion features for tracking. It can handle occlusion to certain extent, and can work well on long duration video sequences. However, when objects are tracked in aerial images taken from platforms like unmanned air vehicle, the problems of frequent pose change, scale and illumination variations arise adding to low resolution, noise and jitter introduced by motion of the camera.  The proposed algorithm incorporates compensation for the camera movement, algorithmic modifications in combining appearance and motion cues for detection and tracking of multiple objects and enhancements in the form of inter object distance measure for improved performance of the tracker when there are many identical objects in proximity. This algorithm has been tested on a large number of aerial sequences including benchmark videos, TLD dataset and many classified unmanned air vehicle sequences and has shown better performance in comparison to TLD.</p><p> </p>


2007 ◽  
Vol 20 (1) ◽  
pp. 23-34 ◽  
Author(s):  
B. Majidi ◽  
A. Bab-Hadiashar

2021 ◽  
Author(s):  
Qianhao Wang ◽  
Yuman Gao ◽  
Jialin Ji ◽  
Chao Xu ◽  
Fei Gao

Sensors ◽  
2020 ◽  
Vol 20 (2) ◽  
pp. 547
Author(s):  
Abu Md Niamul Taufique ◽  
Breton Minnehan ◽  
Andreas Savakis

In recent years, deep learning-based visual object trackers have achieved state-of-the-art performance on several visual object tracking benchmarks. However, most tracking benchmarks are focused on ground level videos, whereas aerial tracking presents a new set of challenges. In this paper, we compare ten trackers based on deep learning techniques on four aerial datasets. We choose top performing trackers utilizing different approaches, specifically tracking by detection, discriminative correlation filters, Siamese networks and reinforcement learning. In our experiments, we use a subset of OTB2015 dataset with aerial style videos; the UAV123 dataset without synthetic sequences; the UAV20L dataset, which contains 20 long sequences; and DTB70 dataset as our benchmark datasets. We compare the advantages and disadvantages of different trackers in different tracking situations encountered in aerial data. Our findings indicate that the trackers perform significantly worse in aerial datasets compared to standard ground level videos. We attribute this effect to smaller target size, camera motion, significant camera rotation with respect to the target, out of view movement, and clutter in the form of occlusions or similar looking distractors near tracked object.


Sign in / Sign up

Export Citation Format

Share Document