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

Table 1

Accuracy of the designed approach using depth database.

Activities	BD	JK	PJU	RUN	SMO	SKI	WAK	1HW	2HW	JU	CL	BOX	SS
BD	99	0	0	0	1	0	0	0	0	0	0	0	0
JK	0	100	0	0	0	0	0	0	0	0	0	0	0
PJU	0	0	98	0	0	1	0	0	0	1	0	0	0
RUN	0	0	0	99	0	0	0	0	1	0	0	0	0
SMO	0	0	1	0	97	0	1	0	0	0	1	0	0
SKI	0	0	0	0	0	99	0	0	0	1	0	0	0
WAK	0	0	1	0	2	0	96	0	1	0	0	1	0
1HW	0	0	0	0	0	0	0	100	0	0	0	0	0
2HW	0	1	0	0	0	0	0	0	99	0	0	0	0
JU	1	0	0	0	1	0	1	0	0	97	0	0	0
CL	0	0	1	1	0	0	0	0	0	0	98	0	0
BOX	0	0	0	0	1	1	0	0	1	0	0	97	0
SS	0	0	0	0	0	0	2	0	0	0	0	0	98
Average	98.2%

BD represents bending, JK represents jacking, PJU represents place-jumping, RUN represents running, SDM represents side movement, SKI represents skipping, WAK represents walking, 1HW represents one hand waving, 2HW represents two hand waving, JU represents jumping, CL represents clapping, BOX represents boxing, and SS represents sitting and standing.