An Efficient Feature Selection Method for Video-Based Activity Recognition Systems

Table 10

Accuracy of the designed approach with variable neighborhood search on depth database.

Activities	BD	JK	PJU	RUN	SMO	SKI	WAK	1HW	2HW	JU	CL	BOX	SS
BD	88	2	0	1	1	2	1	0	2	0	2	0	1
JK	0	89	1	2	1	2	0	1	0	2	0	2	0
PJU	1	0	92	1	0	0	2	1	2	0	1	0	0
RUN	0	2	1	91	0	1	1	0	0	1	0	1	2
SMO	2	0	0	1	93	0	0	1	2	0	1	0	0
SKI	0	2	1	0	2	90	1	0	1	2	1	0	1
WAK	1	0	2	2	0	2	87	2	1	0	2	1	0
1HW	2	1	1	2	1	0	2	85	0	2	0	2	2
2HW	1	0	2	1	0	1	0	1	93	0	1	0	0
JU	0	2	1	0	2	0	1	1	0	91	0	1	1
CL	1	0	0	2	0	1	0	0	1	0	94	1	0
BOX	1	2	0	1	2	0	1	2	0	1	0	89	1
SS	2	0	2	1	1	2	0	1	2	1	2	1	85
Average	89.8%