Hand Gesture Recognition Using Modified 1$ and Background Subtraction Algorithms

Table 2

Performance of the proposed algorithm versus some of the previous algorithms.

Paper	Number of postures	Recognition time; sec/frame	Recognition accuracy	Frame resolution	Number of testing images	Scale	Rotation	Light changes	Background
Proposed algorithm	5	0.04	98.60%	320 × 240	500	Variant	Variant	Variant	Cluttered
[10]	15	0.4	94.89%	160 × 120	30	Not discussed	Invariant	Not discussed	wall
[11]	6	0.09–0.11	93.80%	320 × 240	195–221	Not discussed	Not discussed	Not discussed	Cluttered
[12]	3	0.1333	96.20%	640 × 480	130	Invariant	30 degree	Variant	Different
[13]	4	0.03	90.00%	320 × 240	100	Invariant	15 degree	Invariant	White wall
[16]	8	0.066667	96.90%	640 × 480	300	Variant	Invariant	Not discussed	Not discussed
[17]	6	Not discussed	93.40%	100 × 100	57–76	Invariant	Not discussed	Invariant	wall
[17]	6	Not discussed	76.10%	100 × 100	38–58	Invariant	Not discussed	Not discussed	Cluttered
[18]	10	0.017	96.23%	640 × 480; or any size	1000	Invariant	Invariant	Invariant	Cluttered