跳到主要导航 跳到搜索 跳到主要内容

Hazardous Chemical Vehicles Detection Based on Path Fusion and Multi-Scale Slicing

  • Northwestern Polytechnical University Xian

科研成果: 书/报告/会议事项章节 › 会议稿件 › 同行评审

摘要

In order to ensure the safety of highway transportation, hazardous chemical vehicles detection was carried out on vehicles engaged in the transportation of hazardous chemicals. High-speed and diverse scenes of vehicles on the road pose numerous challenges for object detection, as hazardous chemical transport vehicles often appear as small targets within the surveillance field of view. In this paper, we propose a dedicated vehicle detection algorithm based on path fusion and multi-scale slicing methods. This algorithm incorporates a dual path fusion module and a multi-scale image slice detection module to effectively enhance its feature extraction capability and fuse semantic detail information. By conducting experiments on a self-built dataset, we have demonstrated the reliability and superiority of the proposed algorithm over baseline models and mainstream object detection algorithms in terms of detection accuracy and inference speed.

源语言英语
主期刊名Proceedings of 2023 7th Asian Conference on Artificial Intelligence Technology, ACAIT 2023
出版商Institute of Electrical and Electronics Engineers Inc.
页661-666
页数6
ISBN(电子版)9798350359145
DOI
出版状态已出版 - 2023
活动7th Asian Conference on Artificial Intelligence Technology, ACAIT 2023 - Quzhou, 中国
期限: 10 11月 2023 → 12 11月 2023

丛书

姓名Proceedings of 2023 7th Asian Conference on Artificial Intelligence Technology, ACAIT 2023

会议

会议7th Asian Conference on Artificial Intelligence Technology, ACAIT 2023
国家/地区中国
市Quzhou
时期10/11/23 → 12/11/23

联合国可持续发展目标

此成果有助于实现下列可持续发展目标:

  1. 可持续发展目标 12 - 负责任消费和生产
    可持续发展目标 12 负责任消费和生产

学术指纹

探究 'Hazardous Chemical Vehicles Detection Based on Path Fusion and Multi-Scale Slicing' 的科研主题。它们共同构成独一无二的学术指纹。

引用此