Aiming at the problem of detection and recognition of abnormal sound events in public scenes, this paper proposes an algorithm based on machine hearing algorithm to automatically complete the detection and classification of abnormal sound activities. Through real-time monitoring of the scene, template matching is carried out with the list of abnormal events to realize the judgment of abnormal sound activities in public scenes. The feature mapping of multi-dimensional vector space is completed for the speech segments of potential acoustic activity. The feature vector includes not only the own features, but also the features related to the event list template. The SVM algorithm based on Gaussian radial basis function is used to train and test the performance on the self-organized dataset. The results show that the algorithm has a good performance in detecting the accuracy of classification.
Research on Acoustic Anomaly Detection in Public Scene Based on Multi-dimensional Feature Space
Lect. Notes Electrical Eng.
International Conference on Autonomous Unmanned Systems ; 2021 ; Changsha, China September 24, 2021 - September 26, 2021
Proceedings of 2021 International Conference on Autonomous Unmanned Systems (ICAUS 2021) ; Chapter : 121 ; 1214-1224
2022-03-18
11 pages
Article/Chapter (Book)
Electronic Resource
English
Dynamic scene modelling and anomaly detection based on trajectory analysis
Wiley | 2014
|Expressway traffic anomaly detection method based on multi-dimensional matrix portrait
European Patent Office | 2024
|Target Recognition Based on Multi-Dimensional Polarization Feature Space
British Library Conference Proceedings | 1996
|