基于红外立体视觉的行人感知方法研究

王向军; 杨寿常; 陈瑞祥

基于红外立体视觉的行人感知方法研究

王向军^{1, 2,},
杨寿常^{1, 2},
陈瑞祥^{1, 2}

1.
天津大学精密测试技术及仪器国家重点实验室，天津 300072
2.
天津大学微光机电系统技术教育部重点实验室，天津 300072

详细信息

作者简介:
王向军（1955-），男，教授，博士，主要研究方向为光电探测与测量和图像测量与计算机视觉。E-mail：tjuxjw@126.com

中图分类号: TP391.41
计量
- 文章访问数: 215
- HTML全文浏览量: 44
- PDF下载量: 42
- 被引次数: 0
出版历程
- 收稿日期: 2020-09-23
- 修回日期: 2020-12-09
- 刊出日期: 2021-07-01

Pedestrian Perception Method Based on Infrared Stereo Vision

WANG Xiangjun^{1, 2
,},
YANG Shouchang^{1, 2},
CHEN Ruixiang^{1, 2}

1.
State Key Laboratory of Precision Measuring Technology and Instruments, Tianjin University, Tianjin 300072, China
2.
MOEMS Education Ministry Key Laboratory, Tianjin University, Tianjin 300072, China

摘要

摘要: 基于热红外特性，红外立体视觉路况行人感知方法可以在夜间、雾霾环境下有效检测道路场景中的行人等目标，提高驾驶安全性。针对红外图像中纹理细节少，传统稠密双目立体匹配算法效果差的问题，本文首先根据目标在红外图像下的亮度、边缘特征提取感兴趣区域（Region of interest, ROI）；然后在ROI中提取图像特征点并匹配，进而计算原始稀疏深度图；最后根据目标表面深度变化较小的特点，结合ROI和原始深度图估计半稠密深度图。本文搭建了实验系统验证该方法的有效性。实验结果表明，在系统约120°观测视场角内，该方法对行人等目标深度感知相对误差在15 m范围内优于1.5%，30 m范围内优于3%。
- 长波红外 /
- 立体视觉 /
- SURF /
- 深度图 /
- 辅助驾驶
Abstract: Based on the thermal infrared characteristics, the infrared stereo vision pedestrian perception method can effectively detect and measure pedestrians in road scenes at night and hazy environments, with the aim of improving driving safety. Owing to less texture details in infrared images, the traditional dense binocular stereo matching algorithm performs poorly. To solve this problem, the region of interest (ROI) is extracted according to the brightness and edge features of the targets in the infrared image. Then, the image feature points are extracted and matched in the ROI to calculate the original sparse depth map. Finally, according to the small depth difference in the surface of the targets, the semi-dense depth map was estimated by combining the ROI and the original depth map. We designed an experimental system to verify the effectiveness of the proposed method. The experimental results showed that the relative error of the depth perception of pedestrians was better than 1.5% at 15 m and 3% at 30 m in the field of view of approximately 120°.
- long-wave infrared /
- stereo vision /
- SURF /
- depth map /
- assisted driving

HTML全文

图 1 双目测量模型

Figure 1. Binocular measurement model

下载: 全尺寸图片幻灯片

图 2 系统设计方案

Figure 2. System design scheme

下载: 全尺寸图片幻灯片

图 3 大视场红外立体视觉实验系统

Figure 3. Large field of view infrared stereo vision experimental system

下载: 全尺寸图片幻灯片

图 4 红外棋盘格标定板

Figure 4. Infrared chessboard

下载: 全尺寸图片幻灯片

图 5 感兴趣区域提取

Figure 5. Region of interest (ROI) detection

下载: 全尺寸图片幻灯片

图 6 特征点提取及匹配

Figure 6. Feature points extraction and match

下载: 全尺寸图片幻灯片

图 7 深度图（局部）

Figure 7. Depth map (local)

下载: 全尺寸图片幻灯片

图 8 路面测试结果

Figure 8. Results of road test

下载: 全尺寸图片幻灯片

图 9 2×2红外棋盘格

Figure 9. 2×2 infrared chessboard

下载: 全尺寸图片幻灯片

图 10 左视双目视场目标参考位置

Figure 10. Reference position of targets in left binocular field of view

下载: 全尺寸图片幻灯片

图 11 测距实验结果

Figure 11. Results of distance measurement

下载: 全尺寸图片幻灯片

参考文献(11)

[1]	World Health Organization. The Top 10 Causes of Death[DB/OL]. [2018-05-24]https://www.who.int/news-room/fact-sheets/detail/the-top-10-causes-of-death.
[2]	Gade R, Moeslund T B. Thermal Cameras and Applications: a Survey[J]. Machine Vision and Applications, 2014, 25(1): 245-262. doi: 10.1007/s00138-013-0570-5
[3]	石永彪, 张湧. 车载红外夜视技术发展研究综述[J]. 红外技术, 2019, 41(6): 504-510. https://www.cnki.com.cn/Article/CJFDTOTAL-HWJS201906002.htm SHI Yongbiao, ZHANG Yong. Survey on development about vehicular infrared night vision technology[J]. Infrared Technology, 2019, 41(6): 504-510. https://www.cnki.com.cn/Article/CJFDTOTAL-HWJS201906002.htm
[4]	许腾, 黄铁军, 田永鸿. 车载视觉系统中的行人检测技术综述[J]. 中国图象图形学报, 2013, 18(4): 359-367. https://www.cnki.com.cn/Article/CJFDTOTAL-ZGTB201304002.htm XU Teng, HUANG Tiejun, TIAN Yonghong. Survey on pedestrian detection technology for on-board vision systems[J]. Journal of Image and Graphics, 2013, 18(4): 359-367. https://www.cnki.com.cn/Article/CJFDTOTAL-ZGTB201304002.htm
[5]	于博, 马书浩, 李红艳, 等. 远红外车载图像实时行人检测与自适应实例分割[J]. 激光与光电子学进展, 2020, 57(2): 293-303. https://www.cnki.com.cn/Article/CJFDTOTAL-JGDJ202002033.htm YU Bo, MA Shuhao, LI Hongyan, et al. Real-time pedestrian detection for far-infrared vehicle images and adaptive instance segmentation[J]. Laser & Optoelectronics Progress, 2020, 57(2): 293-303. https://www.cnki.com.cn/Article/CJFDTOTAL-JGDJ202002033.htm
[6]	DING M, ZHANG X, CHEN W H, et al. Thermal Infrared Pedestrian Tracking via Fusion of Features in Driving Assistance System of Intelligent Vehicles[J]. Proceedings of the Institution of Mechanical Engineers Part G-Journal of Aerospace Engineering, 2019, 233(16): 6089-6103. doi: 10.1177/0954410019890820
[7]	ZHANG W Y, FU X H, LI W. The Intelligent Vehicle Target Recognition Algorithm Based on Target Infrared Features Combined with Lidar[J]. Computer Communications, 2020, 155: 158-165. doi: 10.1016/j.comcom.2020.03.013
[8]	Richard Hartley, Andrew Zisserman. 计算机视觉中的多视图几何[M]. 韦穗, 章权兵, 译. 北京: 机械工业出版社, 2019: 240-242. Richard Hartley, Andrew Zisserman. Multiple View Geometry in Computer Vision[M]. WEI Sui, ZHANG Quanbing. Beijing: China Machine Press, 2019: 240-242.
[9]	ZHANG Z. A flexible new technique for camera calibration[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2000, 22(11): 1330-1334. doi: 10.1109/34.888718
[10]	Bertozzi M, Broggi A, Caraffi C, et al. Pedestrian detection by means of far-infrared stereo vision[J]. Computer Vision and Image Understanding, 2006, 106(2): 194-204. doi: 10.1016/j.cviu.2006.07.016
[11]	Bay H, Ess A, Tuytelaars T, et al. SURF: Speeded Up Robust Features[J]. Computer Vision and Image Understanding, 2008, 110(3): 346-359. doi: 10.1016/j.cviu.2007.09.014