Vision-based detection and distance estimation of micro unmanned aerial vehicles

Download

index.pdf

Date

2015

Author

Gökçe, Fatih

Metadata

Show full item record

Item Usage Stats

221
views

156
downloads

In this thesis, we study visual detection and distance estimation of Micro Unmanned Aerial Vehicles (mUAVs), a crucial problem for (i) intrusion detection of mUAVs in protected environments, (ii) sense and avoid purposes on mUAVs or on other aerial vehicles and (iii) multi-mUAV control scenarios such as environmental monitoring, surveillance and exploration. The problem is challenging since (i) a real-time solution is required, a burden when computational power is limited by the hardware carried by an mUAV, (ii) non-convex structure of the mUAVs causes the bounding box of mUAVs to include very different background patterns, (iii) background patterns from indoor or outdoor are very complex with different characteristics and can include moving objects, (iv) mUAVs tilt and rotate unavoidably resulting in very large changes in their appearances, (v) when the camera is not stationary, motion blur is a problem, and (vi) illumination direction and brightness changes cause different images. We evaluate vision algorithms for this problem, since other sensing modalities limit the environment or the distance between the mUAVs. We test Haar-like features, Local Binary Patterns (LBP) and Histogram of Gradients (HOG) using boosted cascaded classifiers. We also integrate a distance estimation method utilizing geometric cues with Support Vector Regressors. We evaluated each method on indoor and outdoor videos collected systematically and on videos with motion blur. Our experiments show that, using boosted cascaded classifiers with LBP, near real-time detection and distance estimation of mUAVs are possible in about 60 ms indoors (1032x778 resolution) and 150 ms outdoors (1280x720 resolution) per frame, with a detection rate of 0.96 F-Score. However, classifiers of Haar-like features lead to better distance estimation since they position the bounding boxes on mUAVs more accurately. Our time analysis yields that classifiers of HOG train and run faster than the other algorithms.

Subject Keywords

Micro air vehicles., Drone aircraft., Computer vision., Distances

URI

http://etd.lib.metu.edu.tr/upload/12619440/index.pdf
https://hdl.handle.net/11511/25226

Collections

Graduate School of Natural and Applied Sciences, Thesis

Suggestions

OpenMETU
Core

Vision-Based Detection and Distance Estimation of Micro Unmanned Aerial Vehicles Gökçe, Fatih; Üçoluk, Göktürk; Şahin, Erol; Kalkan, Sinan (MDPI, ST ALBAN-ANLAGE 66, CH-4052 BASEL, SWITZERLAND, 2015-9) Detection and distance estimation of micro unmanned aerial vehicles (mUAVs) is crucial for (i) the detection of intruder mUAVs in protected environments; (ii) sense and avoid purposes on mUAVs or on other aerial vehicles and (iii) multi-mUAV control scenarios, such as environmental monitoring, surveillance and exploration. In this article, we evaluate vision algorithms as alternatives for detection and distance estimation of mUAVs, since other sensing modalities entail certain limitations on the environment...
Flow visualization around a flapping-wing micro air vehicle in free flight Del Estal Herrero, Alejandro; Perçin, Mustafa; Karasek, Matej; Van Oudheusden, Bas W. (null; 2018-10-05) Flow visualizations have been performed on a free flying flapping-wing Micro Air Vehicle (MAV), using a large-scale particle image velocimetry (PIV) approach. The PIV method involves the use of helium filled soap bubbles (HFSB) as tracer particles. HFSB scatter light with much higher intensity than regular seeding particles and comparable to that reflected off the flexible flapping wings. This enables flow field visualization to be achieved also close to the flapping wings, in contrast to previous PIV exper...
Flow Visualization around a Flapping-Wing Micro Air Vehicle in Free Flight Using Large-Scale PIV Herrero, Alejandro del Estal; Perçin, Mustafa; Karasek, Matej; van Oudheusden, Bas (2018-12-01) Flow visualizations have been performed on a free flying, flapping-wing micro air vehicle (MAV), using a large-scale particle image velocimetry (PIV) approach. The PIV method involves the use of helium-filled soap bubbles (HFSB) as tracer particles. HFSB scatter light with much higher intensity than regular seeding particles, comparable to that reflected off the flexible flapping wings. This enables flow field visualization to be achieved close to the flapping wings, in contrast to previous PIV experiments ...
Coordinated guidance for multiple UAVs Cakici, Ferit; Ergezer, Halit; Irmak, Ufuk; Leblebicioğlu, Mehmet Kemal (2016-05-01) This paper addresses the path planning problem of multiple unmanned aerial vehicles (UAVs). The paths are planned to maximize collected amount of information from desired regions (DRs), while avoiding forbidden regions (FRs) and reaching the destination. This study focuses on maximizing collected information instead of minimizing total mission time, as in previous studies. The problem is solved by a genetic algorithm (GA) with the proposal of novel evolutionary operators. The initial populations are generat...
Vision-aided landing for fixed wing unmanned aerial vehicle Esin, Engin; Kutay, Ali Türker; Department of Aerospace Engineering (2016) The aim of this thesis is to design an autoland system for fixed wing unmanned aerial vehicle (UAV) to make auto landing by using position information calculated by image processing algorithms. With this ability, even if GPS is not available to be used, UAV still could make a safe automatic landing. Landing autopilot is aimed to keep UAV on a straight line with a constant flight path angle. Therefore, landing autopilot and computer vision methods are studied within the scope of this thesis. Also, to test de...

Citation Formats

F. Gökçe, “Vision-based detection and distance estimation of micro unmanned aerial vehicles,” Ph.D. - Doctoral Program, Middle East Technical University, 2015.