基于多传感器信息融合的车底危险品分类识别算法
DOI:
作者:
作者单位:

河北工业大学

作者简介:

通讯作者:

中图分类号:

TP391

基金项目:

:国家重点研发计划项目(2017YFB1302504);河北省自然科学基金项目(F2018202210)


Dangerous target under vehicle classification and recognition algorithm based on multi-sensor information fusion
Author:
Affiliation:

Hebei University of Technology

Fund Project:

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 文章评论
    摘要:

    当前安保机器人用于侦测车底危险品时,主要以视觉识别为主,受限于光照、背景、暴露时长等诸多条件,识别准确率低,效率低。针对以上问题,文中采用气体传感器及辐射传感器信息作为补充输入,综合图像异物率、辐射强度、气体浓度信息,提出一种基于云模型的改进D-S证据理论和加权平均相结合的融合算法,并设计了车底危险品识别系统软件。通过车底危险品识别装置的测试实验,证明了本文算法能完成车底危险品识别的目标,并且能够对危险品等级准确分类,具有良好的实际应用价值。

    Abstract:

    When the current security robot is used to detect dangerous goods under the vehicle, it is mainly based on visual recognition, limited by many conditions such as light, background, and exposure time. The recognition accuracy is low and the efficiency is low. In view of the above problems, the information of gas sensor and radiation sensor is used as supplementary input, and the information of foreign matter rate, radiation intensity and gas concentration in the image is integrated. A fusion algorithm based on cloud model and combining DS evidence theory and weighted average is proposed and designed. Dangerous goods identification system software under the vehicle. Through the test experiment of the vehicle dangerous goods identification device, it is proved that the algorithm of this paper can complete the goal of vehicle dangerous goods identification, and can classify the dangerous goods accurately, which has good practical application value.

    参考文献
    相似文献
    引证文献
引用本文

高春艳. 基于多传感器信息融合的车底危险品分类识别算法[J]. 科学技术与工程, , ():

复制
分享
文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数:
历史
  • 收稿日期:2020-10-01
  • 最后修改日期:2021-01-31
  • 录用日期:2021-02-16
  • 在线发布日期:
  • 出版日期:
×
《科学技术与工程》影响力指数稳步提升
亟待确认的版面费信息