Learning to Detect Aircraft at Low Resolutions

Petridis, Stavros; Geyer, Christopher; Singh, Sanjiv

doi:10.1007/978-3-540-79547-6_46

Cited by 19 publications

(11 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Furthermore, an automated detection system developed by Defense Research Associates was implemented and flight tested, in which the detection approach is based on target pixel movement (or energy flow) in the image frame (Utt, McCalmont, & Deschenes, 2004, 2005). Recently, the problem of aircraft detection using passive vision has been tackled using (1) a combination of morphological filtering and support vector machine classification (Dey, Geyer, Singh, & Digioia, 2009; Geyer, Dey, & Singh, 2009) and (2) a machine learning approach based on AdaBoost that exploits a modified version of the Viola and Jones object detector (Petridis, Geyer, & Singh, 2008).…”

Section: Introductionmentioning

confidence: 99%

Airborne vision‐based collision‐detection system

Lai

Mejias

Ford

2011

Journal of Field Robotics

106

View full text Add to dashboard Cite

Machine vision represents a particularly attractive solution for sensing and detecting potential collision-course targets due to the relatively low cost, size, weight, and power requirements of vision sensors (as opposed to radar and Traffic Alert and Collision Avoidance System). This paper describes the development and evaluation of a real-time, vision-based collision-detection system suitable for fixed-wing aerial robotics. Using two fixedwing unmanned aerial vehicles (UAVs) to recreate various collision-course scenarios, we were able to capture highly realistic vision (from an onboard camera perspective) of the moments leading up to a collision. This type of image data is extremely scarce and was invaluable in evaluating the detection performance of two candidate target detection approaches. Based on the collected data, our detection approaches were able to detect targets at distances ranging from 400 to about 900 m. These distances (with some assumptions about closing speeds and aircraft trajectories) translate to an advance warning of between 8 and 10 s ahead of impact, which approaches the 12.5-s response time recommended for human pilots. We overcame the challenge of achieving real-time computational speeds by exploiting the parallel processing architectures of graphics processing units (GPUs) found on commercial-off-the-shelf graphics devices. Our chosen GPU device suitable for integration onto UAV platforms can be expected to handle real-time processing of 1,024 × 768 pixel image frames at a rate of approximately 30 Hz. Flight trials using manned Cessna aircraft in which all processing is performed onboard will be conducted in the near future, followed by further experiments with fully autonomous UAV platforms. C

show abstract

Section: Introductionmentioning

confidence: 99%

Airborne vision‐based collision‐detection system

Lai

Mejias

Ford

2011

Journal of Field Robotics

106

View full text Add to dashboard Cite

show abstract

“…There are also studies for detecting aircraft via vision [ 55 , 56 , 57 ]. Although we include mainly the literature proposed for mUAVs in this section, these studies are noteworthy, since they are potentially useful for mUAVs, as long as the size, weight and power (SWaP) constraints of mUAVs are complied with.…”

Section: Related Studiesmentioning

confidence: 99%

“…Although we include mainly the literature proposed for mUAVs in this section, these studies are noteworthy, since they are potentially useful for mUAVs, as long as the size, weight and power (SWaP) constraints of mUAVs are complied with. In [ 55 ], aircraft detection under the presence of heavily cluttered background patterns is studied for collision avoidance purposes. They applied a modified version of boosted cascaded classifiers using Haar-like features for detection.…”

Section: Related Studiesmentioning

confidence: 99%

Vision-Based Detection and Distance Estimation of Micro Unmanned Aerial Vehicles

Gökçe

Üçoluk

Şahi̇n

et al. 2015

Sensors

View full text Add to dashboard Cite

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 or on the distance. For this purpose, we test Haar-like features, histogram of gradients (HOG) and local binary patterns (LBP) using cascades of boosted classifiers. Cascaded boosted classifiers allow fast processing by performing detection tests at multiple stages, where only candidates passing earlier simple stages are processed at the preceding more complex stages. We also integrate a distance estimation method with our system utilizing geometric cues with support vector regressors. We evaluated each method on indoor and outdoor videos that are collected in a systematic way and also on videos having 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 (1032×778 resolution) and 150 ms outdoors (1280×720 resolution) per frame, with a detection rate of 0.96 F-score. However, the cascaded classifiers using Haar-like features lead to better distance estimation since they can position the bounding boxes on mUAVs more accurately. On the other hand, our time analysis yields that the cascaded classifiers using HOG train and run faster than the other algorithms.

show abstract

“…Furthermore, it is well known that morphology techniques are computationally intensive requiring parallelised implementation or dedicated hardware to meet real-time requirements. Finally, morphology approaches are noted as returning a large number of false positives [15,16,17] with computationally intensive techniques used to post-process the detections.…”

Section: Algorithm Backgroundmentioning

confidence: 99%

Real-Time Optical Aircraft Detection

Revell

Buchanan

Dixon

2012

Measurement and Control

View full text Add to dashboard Cite

The routine use of Uninhabited Air Systems (UAS) in non-segregated airspace presents many technological and regulatory challenges to the aerospace industry.This paper gives an overview of the collision avoidance problem facing UAS designers and provides a discussion of the current optical aircraft detection and tracking approaches. In addition, the paper describes several important challenges for a computer vision based aircraft detection approach that is designed to form part of an optical UAS collision avoidance system. Some initial flight trial results are presented from our prototype and conclusions drawn.

show abstract

Learning to Detect Aircraft at Low Resolutions

Cited by 19 publications

References 11 publications

Airborne vision‐based collision‐detection system

Airborne vision‐based collision‐detection system

Vision-Based Detection and Distance Estimation of Micro Unmanned Aerial Vehicles

Real-Time Optical Aircraft Detection

Contact Info

Product

Resources

About