Towards an Automatic Aircraft Wreckage Detection Using A Monocular Camera of UAV

Accidents occur suddenly and cannot be avoided either in the air, land, and water. The first step taken by the rescue team is to determine the location of the accident. Crash debris must be discovered even though it is on challenging terrains, such as a mountain, forest, or sea. In practice, a large number of casualties is caused by the delay in handling victims who are in unknown locations. Exploration is hampered due to the vastness of the search area, lack of technology, and the terrain that is difficult to reach by rescue teams. Therefore, in this paper, a wreckage aircraft detection system will be studied using only visual information from air sensing. A single monocular camera mounted on a UAV is employed in the system. Haar-like features and cascaded classifiers that have been widely used for object detection is studied due to its efficiency and practical implementation for most embedded platform attached in UAVs. We believe that this research will help accelerate finding accident sites so that the process of handling victims becomes more useful for the search and rescue team.

[1]  Gary R. Bradski,et al.  Learning OpenCV - computer vision with the OpenCV library: software that sees , 2008 .

[2]  Endah Suryawati Ningrum,et al.  Aksara jawa text detection in scene images using convolutional neural network , 2017, 2017 International Electronics Symposium on Knowledge Creation and Intelligent Computing (IES-KCIC).

[3]  Yoav Freund,et al.  A decision-theoretic generalization of on-line learning and an application to boosting , 1995, EuroCOLT.

[4]  Naixue Xiong,et al.  Aircraft detection in remote sensing images based on saliency and convolution neural network , 2018, EURASIP J. Wirel. Commun. Netw..

[5]  W. Marsden.  I and J , 2012 .

[6]  David Hyunchul Shim,et al.  Aircraft Detection using Deep Convolutional Neural Network in Small Unmanned Aircraft Systems , 2018 .

[7]  Jason J. Ford,et al.  Learning to Detect Aircraft for Long-Range Vision-Based Sense-and-Avoid Systems , 2018, IEEE Robotics and Automation Letters.

[8]  Timothy Molloy,et al.  Detection of aircraft below the horizon for vision‐based detect and avoid in unmanned aircraft systems , 2017, J. Field Robotics.

[9]  Jemal H. Abawajy,et al.  Text Detection in Low Resolution Scene Images Using Convolutional Neural Network , 2016, SCDM.

[10]  Li Fei-Fei,et al.  ImageNet: A large-scale hierarchical image database , 2009, CVPR.

[11]  Paul A. Viola,et al.  Rapid object detection using a boosted cascade of simple features , 2001, Proceedings of the 2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. CVPR 2001.

[12]  Timothy Molloy,et al.  Below Horizon Aircraft Detection Using Deep Learning for Vision-Based Sense and Avoid , 2019, 2019 International Conference on Unmanned Aircraft Systems (ICUAS).

[13]  Anhar Risnumawan,et al.  Deep multilayer network for automatic targeting system of gun turret , 2017, 2017 International Electronics Symposium on Engineering Technology and Applications (IES-ETA).

[14]  Yoav Freund,et al.  A Short Introduction to Boosting , 1999 .

[15]  Anhar Risnumawan,et al.  From concrete to abstract: Multilayer neural networks for disaster victims detection , 2016, 2016 International Electronics Symposium (IES).

[16]  Zaqiatud Darojah,et al.  Deep Features Representation for Automatic Targeting System of Gun Turret , 2018, 2018 International Electronics Symposium on Engineering Technology and Applications (IES-ETA).