Abnormal event detection in indoor environment based on acoustic signal processing

Abdrakhmanov R. Tolep A. Kozhamkulova Z. Narbekov N. Dossanov N. Yeskarayeva B.
31 May 2021 Little Lion Scientific

Journal of Theoretical and Applied Information Technology
2021 #99 Issue 10 2192 - 2205 pp.

Alert the public about emergencies is to bring to public alerts and emergency information on dangers arising from the threat or occurrence of emergency situations of natural and technogenic character, as well as the conduct of hostilities or owing to these actions, the rules of behavior of the population and the need for protection activities. The aim of the work is to develop a method for detecting the sounds of critical situations in the sound stream. In this paper, the term critical situation is understood as an event, the characteristic sound signs of which can speak of acoustic artifacts (a shot, a scream, a glass strike, an explosion, a siren, etc.). The developed method allows you to classify events into two groups: Normal (for example, street noise) and critical situations (for example, an explosion, a scream, a shot). To determine events, machine learning is used, namely the Support Vector Machine method, which solves classification and regression problems by constructing a nonlinear plane separating the solutions. SVM has a fairly wide application in data classification and shows good results in event detection problems. As part of the work, the minimum set of features for the machine learning model was determined, small training and test samples were formed, and a method was developed that classifies normal and abnormal events.

Acoustic Signals , Audio Event , Impulsive Sounds , Machine Learning , Signal Processing

Text of the article Перейти на текст статьи

Khoja Akhmet Yassawi International, Kazakh-Turkish University, Turkistan, Kazakhstan
Almaty University of Power Engineering and Telecommunication, Almaty, Kazakhstan
Caspian University of Technology and Engineering named after Sh.Yessenov, Aktau, Kazakhstan
Turkestan Regional Branch, Academy of Public Administration under the President of the Republic of Kazakhstan, Turkistan, Kazakhstan
Tenaga National University, Kajang, Selangor, 43000, Malaysia

Khoja Akhmet Yassawi International
Almaty University of Power Engineering and Telecommunication
Caspian University of Technology and Engineering named after Sh.Yessenov
Turkestan Regional Branch
Tenaga National University

10 лет помогаем публиковать статьи Международный издатель

Книга Публикация научной статьи Волощук 2026 Book Publication of a scientific article 2026