A Combined One-Class SVM and Template-Matching Approach for User-Aided Human Fall Detection by Means of Floor Acoustic Features

Table 4

Data used in “Set 2.”

(a) Composition of “Set 2”

Class	Number of occurrences
Human falls	44
Basket	7
Fork	7
Ball	8
Book	7
Bag	8
Chair	7

(b) Templates of “Set 2”

Class	Number of templates
	Clean	Noisy
Basket	55	57
Fork	39	55
Ball	11	52
Book	26	57
Bag	26	56
Chair	86	89
Total	243	366