讲座题目：Finding Flight MH370, is Satellite Remote Sensing the Solution?
内容简介： Currently, there is no system tracking aircraft which provides global coverage and is independent of aircraft’s onboard instruments. The implications of this were highlighted in 2014 with the disappearance of Malaysian Airlines Flight MH370, which sparked worldwide outrage and concern that it is possible to lose a Boeing 777 in the 21st century. The subsequent failed three-year search mission - the largest in aviation history in terms of scale, complexity, time and cost - has been unsuccessful thus far, due to the remote and expansive defined search zone in the southern Indian Ocean.
This study explored the ability of satellite imagery to locate aircraft over regions of open water, thus providing a new way to independently track aircraft with currently existing technology. Landsat 8 and Sentinel-2 imagery was obtained over three coastal airports, along with accurate flight data from the airport for analysis verification. Initially, a visual investigation was undertaken on the true colour imagery to locate all aircraft present, followed by an inspection of the infrared wavelengths to determine their applicability for aircraft detection. Three image analysis techniques were applied to the images; unsupervised classification, Principal Component Analysis (PCA) and change detection. These were chosen to attempt to highlight aircraft locations and differentiate from other objects such as ships and cloud cover, as a possible way to automate aircraft detection.
The unsupervised classification method was found least useful, and the PCA and change detection worked to varying degrees of success with each having their own drawbacks. The major limitations found during this process were; image spatial resolution, cloud cover and instrument sensitivity. Further research with higher spatial resolution imagery and different spectral bands is recommended in this area, to provide a viable, more efficient and economic method of global aircraft tracking as an initial search technique.
个人简介： Prof. Linlin Ge is currently an Associate Professor of remote sensing and earth observation in the School of Civil and Environmental Engineering at the University of New South Wales (UNSW), Sydney, Australia. He has been studying earth observation techniques since 1985. Among many other prestigious awards, A/Professor Ge and his team’s work to support the Sichuan Earthquake rescue effort won them the highly prestigious JK Barrie Award for Overall Excellence at the 2008 Asia-Pacific Spatial Excellence Awards. He was named "NSW Scientist of the Year 2009" in the category Physics, Earth Sciences, Chemistry and Astronomy for his work in near real-time satellite remote sensing.He is a Representative on the National Committee of the Remote Sensing and Photogrammetry Commission and foundation member of the Surveying and Spatial Sciences Institute. He is a Member of the IEEE (the Institute of Electrical and Electronics Engineers, Inc.) and a Life Member of the AGU (the American Geophysical Union). He is also a Principal Investigator for satellite missions such as Envisat, ALOS, COSMO-SkyMed, Radarsat-2 and ALOS-2.