On classifying silhouettes in adverse conditions

Sanderson, C. and Gibbins, D. (2004). On classifying silhouettes in adverse conditions. In: Proceedings of the 2004 International Conference on Intelligent Sensors, Sensor Networks and Information Processing (ISSNIP 2004). International Conference on Intelligent Sensors, Sensor Networks and Information Processing (ISSNIP 2004), Grand Hyatt, Melbourne, Australia, (173-178). 14-17 December 2004. doi:10.1109/ISSNIP.2004.1417457


Author Sanderson, C.
Gibbins, D.
Title of paper On classifying silhouettes in adverse conditions
Conference name International Conference on Intelligent Sensors, Sensor Networks and Information Processing (ISSNIP 2004)
Conference location Grand Hyatt, Melbourne, Australia
Conference dates 14-17 December 2004
Convener University of Melbourne
Proceedings title Proceedings of the 2004 International Conference on Intelligent Sensors, Sensor Networks and Information Processing (ISSNIP 2004)
Journal name Proceedings of the 2004 Intelligent Sensors, Sensor Networks
Place of Publication Piscataway, NJ, U.S.A.
Publisher IEEE - Institute for Electrical and Electronic Engineers Inc.
Publication Year 2004
Year available 2004
Sub-type Fully published paper
DOI 10.1109/ISSNIP.2004.1417457
ISBN 0-7803-8894-1
Start page 173
End page 178
Total pages 6
Language eng
Abstract/Summary We compare the performance of holistic and local feature approaches for the purpose of classifying silhouettes in adverse conditions (i.e. occlusions by other silhouettes, noise and imperfect localization by a region of interest algorithm, resulting in clipping and scale changes). Holistic feature extractors based on Hu's moment invariants and principal component analysis (PCA) are coupled with a classifier based on Gaussian densities, while a local feature extractor based on the 2D Hadamard transform (HT) is coupled with a Gaussian mixture model (GMM) based classifier. Experiments show that the HT/GMM approach is relatively robust to clipping, scale changes and occlusions; however in its current form it is highly sensitive to noise. The results further show that the moment based approach achieves relatively poor performance in advantageous conditions and is easily affected by clipping and occlusions: the PCA based approach is highly affected by scale changes and clipping, while being relatively robust to occlusions and noise.
Subjects 080109 Pattern Recognition and Data Mining
080106 Image Processing
080104 Computer Vision
010401 Applied Statistics
Keyword Gaussian mixture model (GMM)
Holistic feature extractors
Local feature extractor
Occlusions
Q-Index Code E1
Additional Notes This conference was held in conjunction with the DEST International Workshop on Signal Processing for Sensor Networks

 
Versions
Version Filter Type
Citation counts: TR Web of Science Citation Count  Cited 0 times in Thomson Reuters Web of Science Article
Google Scholar Search Google Scholar
Created: Wed, 01 Apr 2009, 13:30:56 EST by Mary-Anne Marrington on behalf of School of Information Technol and Elec Engineering