Abstract
This paper addresses the problem of abnormal acoustic event detection in indoor surveillance systems related to safety and security. The proposed concept event detector determines if the acoustic state is either normal or abnormal from accumulated series of acoustic signals using MFCC and deltas coefficients as acoustic feature vectors and a multiclass Adaboost based acoustic context classifier. A novel concept of adopting an exponential criterion and weighted least square solution to boost binary weak classifiers is proposed here for performance and speed improvements over the conventional and prominent GMM based classifiers.
Original language | English |
---|---|
Article number | 6626247 |
Pages (from-to) | 615-622 |
Number of pages | 8 |
Journal | IEEE Transactions on Consumer Electronics |
Volume | 59 |
Issue number | 3 |
DOIs | |
Publication status | Published - 2013 Oct 31 |
Fingerprint
Keywords
- Abnormal event detection
- acoustic signalclassification
- context awareness
- multiclass Adaboost
ASJC Scopus subject areas
- Electrical and Electronic Engineering
- Media Technology
Cite this
Acoustic signal based abnormal event detection in indoor environment using multiclass adaboost. / Lee, Younghyun; Han, David K.; Ko, Hanseok.
In: IEEE Transactions on Consumer Electronics, Vol. 59, No. 3, 6626247, 31.10.2013, p. 615-622.Research output: Contribution to journal › Article
}
TY - JOUR
T1 - Acoustic signal based abnormal event detection in indoor environment using multiclass adaboost
AU - Lee, Younghyun
AU - Han, David K.
AU - Ko, Hanseok
PY - 2013/10/31
Y1 - 2013/10/31
N2 - This paper addresses the problem of abnormal acoustic event detection in indoor surveillance systems related to safety and security. The proposed concept event detector determines if the acoustic state is either normal or abnormal from accumulated series of acoustic signals using MFCC and deltas coefficients as acoustic feature vectors and a multiclass Adaboost based acoustic context classifier. A novel concept of adopting an exponential criterion and weighted least square solution to boost binary weak classifiers is proposed here for performance and speed improvements over the conventional and prominent GMM based classifiers.
AB - This paper addresses the problem of abnormal acoustic event detection in indoor surveillance systems related to safety and security. The proposed concept event detector determines if the acoustic state is either normal or abnormal from accumulated series of acoustic signals using MFCC and deltas coefficients as acoustic feature vectors and a multiclass Adaboost based acoustic context classifier. A novel concept of adopting an exponential criterion and weighted least square solution to boost binary weak classifiers is proposed here for performance and speed improvements over the conventional and prominent GMM based classifiers.
KW - Abnormal event detection
KW - acoustic signalclassification
KW - context awareness
KW - multiclass Adaboost
UR - http://www.scopus.com/inward/record.url?scp=84886564908&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84886564908&partnerID=8YFLogxK
U2 - 10.1109/TCE.2013.6626247
DO - 10.1109/TCE.2013.6626247
M3 - Article
AN - SCOPUS:84886564908
VL - 59
SP - 615
EP - 622
JO - IEEE Transactions on Consumer Electronics
JF - IEEE Transactions on Consumer Electronics
SN - 0098-3063
IS - 3
M1 - 6626247
ER -