SIA OpenIR  > 智能检测与装备研究室
基于雷达与视觉信息融合的低地板有轨电车行人识别方法
Alternative TitleRadar and visual information fusion based low-floor rail car passenger recognition method
杜劲松; 王伟; 白珈俊; 田星; 高洁
Department智能检测与装备研究室
Rights Holder中国科学院沈阳自动化研究所
Patent Agent21002 沈阳科苑专利商标代理有限公司
Country中国
Subtype发明授权
Status有权
Abstract本发明涉及一种基于雷达与视觉信息融合的低地板有轨电车行人识别方法,雷达通过发射和接收信号,得到目标所在二维图像中的目标点的坐标;生成待检测区域;训练似物行人检测模型,将待检测区域分成若干个大小相同的子窗口,通过滑动窗口遍历每个子窗口,计算二值化规范梯度特征与似物行人检测模型的二值化规范梯度特征的相似度,并与设定的相似度阈值进行比较,将包含待检测区域的子窗口进行聚类融合,得到一个包含完整待检测目标的窗口;计算窗口的方向梯度直方图特征。本发明采用毫米波雷达信息获取方式,直接获取待检测障碍物的距离、速度信息;采用毫米波雷达检测信息,有利于快速的在图像上分割出感兴趣区域,减小图像识别时的搜索区域。
Other AbstractThe method involves containing a to-be-detected object in a window if similarity is greater than a preset similarity threshold value. Clustering process is performed to a to-be-detected region in a window to obtain a to-be-detected intact target in the window. Directional gradient histogram feature of the intact target is calculated. Judgment is made to check whether the object is a pedestrian by utilizing a support vector machine according to a pedestrian detection model. An alarm is provided if the object is the pedestrian. The alarm is not provided if the object is not the pedestrian.
PCT Attributes
Application Date2017-10-18
2018-02-23
Date Available2020-02-21
Application NumberCN201710970932.2
Open (Notice) NumberCN107729843B
Language中文
Contribution Rank1
Document Type专利
Identifierhttp://ir.sia.cn/handle/173321/26397
Collection智能检测与装备研究室
Affiliation中国科学院沈阳自动化研究所
Recommended Citation
GB/T 7714
杜劲松,王伟,白珈俊,等. 基于雷达与视觉信息融合的低地板有轨电车行人识别方法[P]. 2018-02-23.
Files in This Item: Download All
File Name/Size DocType Version Access License
CN201710970932.2授权.p(547KB)专利 开放获取CC BY-NC-SAView Download
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[杜劲松]'s Articles
[王伟]'s Articles
[白珈俊]'s Articles
Baidu academic
Similar articles in Baidu academic
[杜劲松]'s Articles
[王伟]'s Articles
[白珈俊]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[杜劲松]'s Articles
[王伟]'s Articles
[白珈俊]'s Articles
Terms of Use
No data!
Social Bookmark/Share
File name: CN201710970932.2授权.pdf
Format: Adobe PDF
This file does not support browsing at this time
All comments (0)
No comment.
 

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.