A Computer Vision Based Sea Search Method Using Kalman Filter and CAMSHIFT
Refereed conference paper presented and published in conference proceedings

Full Text

Times Cited
Web of Science4WOS source URL (as at 10/05/2021) Click here for the latest count

Other information
AbstractWe develop a computer vision based system to assist the aircrew in search and rescue operations at sea. The Government flying service of Hong Kong sends out fixed wing planes to search for survivors after a maritime accident is reported. The search usually lasts for a few hours and is currently conducted by the naked eyes to search for orange colored life boats from a height of 500 ft. It is a tiresome task and true targets may be missed because of human errors. Therefore we propose to use computer vision techniques to tackle the problem. We suggest that using tracking techniques such as the Kalman combined with the CAMSHIFT algorithm can generate good results. Since the traditional CAMSHIFT algorithm fails in tracking small targets especially when they are moving fast, we propose a new extended CAMSHIFT algorithm with changing window sizes based on the speed of the object being tracked. Additionally, we sample the data in the window with a certain Gaussian distribution to make the tracking more accurate. The experimental result shows that this approach is feasible and is able to track small live boat objects correctly at sea.
All Author(s) ListZhang Z, Wong KH
Name of ConferenceInternational Conference on Technological Advances in Electrical, Electronics and Computer Engineering (TAEECE)
Start Date of Conference09/05/2013
End Date of Conference11/05/2013
Place of ConferenceKonya
Country/Region of ConferenceTurkey
Detailed descriptionTo ORKTS: Keyword:
4. Camshift
Pages188 - 193
LanguagesEnglish-United Kingdom
KeywordsCAMSHIFT; Computer Vision; Kalman Filter; Sea Search
Web of Science Subject CategoriesComputer Science; Computer Science, Theory & Methods; Engineering; Engineering, Electrical & Electronic

Last updated on 2021-11-05 at 00:19