Hand Gesture Recognition Algorithm Using SVM and HOG Model for Control of Robotic System

Table 6

Results of comparison with other hand gesture detection systems.

Method	Training data (frame)	Platform	Accuracy (%)	Detection frame (seconds)	Hardware
Combination edge detectio [17]	3154	CPU	82	From 10 to 15	CPU (i5, 2.3 GHz, 16 GB RAM)
HOG characters and SVM [18]	1000	CPU	91	N/A	N/A
Boosted classifiers and active learning [19]	300	CPU	70	0.089	Pentium 4, 3.2 GHz 1 GB RAM
Our proposal	1000	CPU	90	0.15	CPU (i5, 2 GHz, 4 Gb RAM)