Extraction of Temporal Features on Fibonacci Space for Audio Based Vehicle Classification

• Published in: Recent Trends in Image Processing and Pattern Recognition Vol. 1576; pp. 338 - 345
• Main Authors: Sinha, Amandeep, Kumar, S. Hemanth, Prabhakar, Gudmalwar Ashishkumar, Rao, Ch V. Rama
• Format: Book Chapter
• Language: English
• Published: Switzerland Springer International Publishing AG 2022; Springer International Publishing
• Series: Communications in Computer and Information Science
• Subjects: CNN; Pattern recognition; Temporal features; Vehicle sound
• ISBN: 3031070046; 9783031070044
• ISSN: 1865-0929; 1865-0937
• DOI: 10.1007/978-3-031-07005-1_29

In this study we address automatic vehicle and engine identification based on audio information. This information be based on many factors, such as vehicle type, tires, speed, wear and tear of vehicles, as well as type of road. We have decided a feature set for discriminating pairs of classes. Feature set include Fibonacci feature space, entropy, skewness and kurtosis. The audio information collected are real time on-road recordings. There are four classes of vehicle sounds. The paper also shows problems related to vehicles classification. Classification on audio-based engine and vehicle type identification are proposed and conclusions are shown.