A machine learning approach for posture recognition based on simplified shock graph

Authors:
Nooritawati Md Tahir;Aini Hussain;Salina Abdul Samad;Hafizah Husain
Affiliations:
Faculty of Electrical Engineering, Universiti Teknologi Mara, Selangor, Malaysia;Dept. of Electrical, Electronics and Systems, Faculty of Engineering, Universiti Kebangsaan Malaysia, Selangor, Malaysia;Dept. of Electrical, Electronics and Systems, Faculty of Engineering, Universiti Kebangsaan Malaysia, Selangor, Malaysia;Dept. of Electrical, Electronics and Systems, Faculty of Engineering, Universiti Kebangsaan Malaysia, Selangor, Malaysia
Venue:
WAV'09 Proceedings of the 3rd WSEAS international symposium on Wavelets theory and applications in applied mathematics, signal processing & modern science
Year:
2009

Citing 7
Cited 1

Shock Graphs and Shape Matching

International Journal of Computer Vision
A tree-edit-distance algorithm for comparing simple, closed shapes

SODA '00 Proceedings of the eleventh annual ACM-SIAM symposium on Discrete algorithms
Shape matching using edit-distance: an implementation

SODA '01 Proceedings of the twelfth annual ACM-SIAM symposium on Discrete algorithms
Neural Networks for Pattern Recognition

Neural Networks for Pattern Recognition
A Tutorial on Support Vector Machines for Pattern Recognition

Data Mining and Knowledge Discovery
Shock-Based Indexing into Large Shape Databases

ECCV '02 Proceedings of the 7th European Conference on Computer Vision-Part III
A shock grammar for recognition

CVPR '96 Proceedings of the 1996 Conference on Computer Vision and Pattern Recognition (CVPR '96)

Face recognition as an airport and seaport security tool

WSEAS Transactions on Information Science and Applications

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this paper, posture classification using Simplified Shock Graph as feature vectors based on two machine learning techniques namely Artificial Neural Network along with Support Vector Machine are investigated. Initial results showed that both classifiers are able to classify the four main postures with high recognition rate. Moreover, the tremendous performance of Support Vector Machine (SVM) as classifier is confirmed based on the Kappa Score calculated. Initial findings have proven that SSG is apt as feature vectors for posture recognition whilst ANN and SVM were apposite to perform the classification task.