Frame-Based Facial Expression Recognition Using Geometrical Features

Table 3

The confusion matrix obtained by applying our approach on CK+ database. The first row in each expression represents results using SVM classifier. The other row shows the results using KNN classifier.

	An	Di	Fe	Ha	Sa	Su	Ne
An	68.88	2.222	0.00	0.00	6.666	0.00	22.22
	46.66	8.888	2.222	0.00	15.55	0.00	26.66
Di	1.694	62.71	1.694	0.00	1.69 4	0.00	32.20
	8.474	55.93	10.16	1.694	1.694	0	22.03
Fe	0.00	4.00	64.0	4.0	0.00	0.00	28.0
	0.00	8.00	68.00	4.00	12.00	0.00	8.00
Ha	0.00	0.00	0.00	98.55	0.00	0.00	1.449
	0.00	1.449	4.347	92.75	0.00	0.00	1.449
Sa	3.571	0.00	3.571	0.00	32.14	0.00	60.71
	0.00	10.71	14.28	0.00	28.57	0.00	46.42
Su	0.00	1.25	0.00	0.00	0.00	98.75	0.00
	0.00	1.25	0.00	0.00	0.00	98.75	0.00
Ne	2.03	4.568	1.015	0.00	1.522	0.507	90.35
	5.076	6.091	3.045	0.00	5.583	1.015	79.18