MVD, MVDA
收藏资源简介:
本研究介绍了两个用于声学车辆类型分类的新数据集:MVD(移动车辆检测)和MVDA(移动车辆检测增强版)。MVD包含4229个音频样本,涵盖卡车、汽车、摩托车和无车辆类别,记录于印度德里-NCR地区的四个不同城市地点。MVDA是MVD的扩展版本,包含16,916个样本,通过声音变异和音频增强提高模型的泛化能力。这些数据集旨在支持声学交通监控算法的发展,解决交通管理和监控中的实际问题。
This study introduces two novel datasets for acoustic vehicle type classification: MVD (Mobile Vehicle Detection) and MVDA (Mobile Vehicle Detection Augmented). MVD consists of 4,229 audio samples covering four categories: truck, car, motorcycle, and no vehicle, which were recorded at four distinct urban locations in the Delhi-NCR region of India. MVDA, an extended version of MVD, contains 16,916 audio samples and enhances model generalization through sound variation and audio augmentation. These datasets are designed to support the development of acoustic traffic monitoring algorithms and address practical issues in traffic management and monitoring.




