Share Email Print

Proceedings Paper

Challenges in object detection in above-water imagery
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

Many existing methods of object detection, including edge detection, blob detection, and background subtraction (implemented in libraries such as OpenCV) have proven to be enormously successful when applied to many types of video datasets. However, detecting objects over water presents challenges that are unique and not easily accommodated for by pre-existing algorithms available in popular image processing libraries. In this paper, existing approaches are brie y reviewed, and the challenges encountered in above-water video datasets are highlighted. A recently proposed approach to object detection in radar images - a novel, pixel-intensity statistic based thresholding approach | is then reviewed. In this paper, this approach has been successfully applied to EO/IR datasets as well, extending the implementation to ensure success when applied onto other types of image datasets.

Paper Details

Date Published: 7 May 2019
PDF: 12 pages
Proc. SPIE 11018, Signal Processing, Sensor/Information Fusion, and Target Recognition XXVIII, 1101818 (7 May 2019); doi: 10.1117/12.2518879
Show Author Affiliations
Sarah Babbitt, Defence Research and Development Canada (Canada)
Tanya Gatsak, Univ. of Waterloo (Canada)
Bhashyam Balaji, Defence Research and Development Canada (Canada)

Published in SPIE Proceedings Vol. 11018:
Signal Processing, Sensor/Information Fusion, and Target Recognition XXVIII
Ivan Kadar; Erik P. Blasch; Lynne L. Grewe, Editor(s)

© SPIE. Terms of Use
Back to Top
Sign in to read the full article
Create a free SPIE account to get access to
premium articles and original research
Forgot your username?