Controlling Embedded Systems Remotely via Internet-of-Things Based on Emotional Recognition

Table 2

HOG and SVM optimized parameters for training and testing with the LS learning algorithm.

Exp. No.	HOG-CS	HOG-bs	Image size	Feature length/image	Training time	Testing time	Accuracy	ConMat
1	11	2	100	864	4.6	1.1	88.6000	809	110
								118	963
2	7	2	100	2808	7.3	4.6	87.9000	801	118
								124	957
3	9	2	100	1440	7.31	3.7	88.4500	808	111
								120	961
4	13	2	100	432	5.3	0.3	87.7	802	117
								129	952
5	15	2	100	360	6.40	0.4	88.20	807	112
								124	957