Through-Wall Radar Classification of Human Posture Using Convolutional Neural Networks

Table 2

Confusion matrices of experimental results for the free-space scenario.

(a)

				Actual
				Empty	Standing				Sitting
					Micro		Macro		Micro		Macro
					@ 2.5 m	@ 5 m	@ 2.5 m	@ 5 m	@ 2.5 m	@ 5 m	@ 2.5 m	@ 5 m
Predicted	Empty			200	1	3	1	1	1	3		3
	Standing	Micro	@2.5 m		198				1
			@5 m			196				1
		Macro	@2.5 m				196				2
			@5 m					196				1
	Sitting	Micro	@2.5 m		1				198
			@5 m			1				196
		MACRO	@2.5 m				3				198
			@5 m					3				196	Total
			Total	200	200	200	200	200	200	200	200	200	1800
			Accuracy	100%	99%	98%	98%	98%	99%	98%	99%	98%	98.56%

(b)

		Actual
		Empty	Standing	Sitting
Predicted	Empty	200	6	7
	Standing	0	786	5
	Sitting	0	8	788	Total
	Total	200	800	800	1800
	Accuracy	100.00%	98.25%	98.50%	98.56%