Lower Order Krawtchouk Moment-Based Feature-Set for Hand Gesture Recognition

Table 3

Recognition accuracies at different classifiers for ISL images taken at different values of and with raw feature-set.

Values of and	Total F.V.	-NN using ED	-NN using MD	SVM			ELM			MLP
				Poly	RBF	PUK	Poly	RBF	Linear
Raw feature-set for ISL database
	9	84.9%	85.3%	82.9%	83.5%	84.9%	66.2%	65.4%	66.1%	84.5%
	18	85%	86.1%	83%	84.3%	85.4%	76%	75.8%	76.4%	85.3%
	27	86.1%	86.1%	84.1%	84.7%	86.4%	84.5%	82.7%	84.2%	86.9%
	36	86.9%	86.3%	84.1%	86.6%	86.8%	85.3%	84.47%	84.9%	86.9%
	45	86.9%	86.9%	86.1%	87.4%	87.9%	88.1%	87.7%	87.9%	87.3%
Normalized feature-set for ISL database
	9	85.5%	85.9%	85.9%	84.8%	85.9%	70.7%	70.2%	70.1%	85.2%
	18	86%	86.5%	87.4%	87.4%	87.5%	80%	79.2%	80.4%	85.8%
	27	87.6%	88.1%	88.5%	87.3%	87.7%	86.2%	87.8%	85.9%	86%
	36	88.2%	88.1%	88.6%	88.2%	88.9%	87.5%	87.9%	86.8%	86.3%
	45	88.2%	88.2%	88.9%	89.2%	90%	89.3%	89.2%	88.9%	87.6%