摘要
随着智能驾驶普及,特种车辆识别成为关键问题。文章针对智能驾驶时代特种车辆识别难题展开研究,首先介绍基于声音识别的车辆级系统框架,然后详细阐述运用TorchAudio库进行声音预处理、特征提取,最后介绍结合 ResNet-18 残差网络进行声音分类的方法,并通过实验验证该方法在特种车辆声音识别上的有效性,为智能驾驶中特种车辆识别提供新途径。
Abstract
With the popularization of intelligent driving, the identification of special vehicles has become a key issue. This article conducts research on the recognition challenges of special vehicles in the era of intelligent driving.Firstly, it introduces the vehicle-level system framework based on sound recognition. Then, it elaborates in detail the methods of sound preprocessing and feature extraction using the TorchAudio library. Finally, it combines the ResNet-18 residual network for sound classification. And the effectiveness of this method in the sound recognition of special vehicles is verified through experiments, providing a new approach for the recognition of special vehicles in intelligent driving.
关键词
智能驾驶 /
特种车辆 /
声音识别 /
Python /
Torchaudio /
ResNet-18
Key words
intelligent driving /
special vehicles /
voice recognition /
Python /
Torchaudio /
ResNet-18
梁石杨, 刘 璐.
基于声音识别的智能驾驶特种车辆识别方法[J]. 汽车电器. 2025, 1(10): 21-23
Liang Shiyang, Liu Lu.
Intelligent Driving Special Vehicle Recognition Method Based on Sound Recognition[J]. AUTO ELECTRIC PARTS. 2025, 1(10): 21-23
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] 中华人民共和国道路交通安全法 [M]. 北京:法律出版社,2005.
[2] Kaiming He,Xiangyu Zhang,Shaoqing Ren,et al.Deep Residual Learning for Image Recognition[C]//Proceedings of the IEEE conference on computer vision and pattern recognition,2015:770-778.
[3] 崔胜民 . 一本书读懂无人驾驶汽车 [M]. 北京:机械工业出版社,2024.
[4] 吴斌 . 汽车内 A2B 音频系统设计研究 [J]. 汽车实用技术,2024(2):121-124.
[5] ADI.SHARC+Single Core High Performance DSP (Up to 1 GHz)[EB/OL].(2023-01-01)[2024-01-01].https://www.analog.com/en/products/sharc-dsps/sharc-single-core-dsps.html.
[6] 白龙,王淑玉,付东辉,等 . 基于 DSP 的 CAN 总线网络设计 [J]. 电子设计工程,2017,25(17):168-171.
[7] 张梦龙,龚勋阳,夏皓伦 .Pandas 和 Matplotlib 在数据分析与可视化中的应用 [J]. 信息技术与信息化,2024(3):150-153.
[8] 魏化永 . 基于 Pydub 的音频混剪系统设计与实现 [J]. 电脑编程技巧与维护,2024(4):45-47.
[9] 陶佰睿,郭琴,苗凤娟,等 . 基于自适应 Mel 滤波器组的MFCC 特征提取的 SOC 设计 [J]. 吉林大学学报 ( 工学版 ),2016,46(S1):388-392.
[10] 张进军 .PyTorch 框架下的复杂场景目标识别方法研究 [J].计算机应用与软件,2024,41(5):240-244.
[11] 左一鹏,陈辉 . 基于 Scikit-Learn 的 SVM 分类器算法优化 [J].计算机技术与发展,2020,30(7):116-120.