Altered engine recognition is of significant research and application prospects since it is a sub-task of environmental sound classification and helps detect altered vehicles for law enforcement. In this work, we propose a simple but effective framework based on a sound embedding network with an attention mechanism to identify noise emitted from vehicles with altered engines. Real-world vehicle sound data were collected at a number of streets with different amounts of traffic in Hong Kong. In particular, we developed a proprietary dataset consisting of the environmental and engine sounds through manual segmentation and annotation, and we found that the attention mechanism can help emphasize the informative segments in the frame-level embeddings outputted by the embedding network. We also demonstrate the effectiveness of the proposed attention mechanism on the ESC-50 and UrbanSound8K datasets. The experimental results show that our method achieves remarkable performance on UrbanSound8K and outperforms the baseline on ESC-50 by more than 10%.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Deep Segment-Attentive Network for Altered-Engine Recognition


    Contributors:


    Publication date :

    2023-09-24


    Size :

    1263154 byte





    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English



    Attentive Billboards

    Haritaoglu, I. / Flickner, M. / IEEE | British Library Conference Proceedings | 2001


    Attentive billboards

    Haritaoglu, I. / Flickner, M. | IEEE | 2001


    Pre-Attentive and Attentive Detection of Humans in Wide-Field Scenes

    Elder, J. H. / Prince, S. J. / Hou, Y. et al. | British Library Online Contents | 2007


    Attentive Systems: A Survey

    Nguyen, T. V. / Zhao, Q. / Yan, S. | British Library Online Contents | 2018


    LiDAR Imaging-based Attentive Perception

    Tsiourva, Maria / Papachristos, Christos | IEEE | 2020