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

Table 3

Data used in “Set 1.”

(a) Composition of “Set 1”

Class	Number of occurrences
Human falls	44
Human activity	15
Music	29

(b) Templates of “Set 1”

Class	Number of templates
	Clean	Noisy
Human activity	13	11
Music	8	16
Total	21	27