基于行人关键点的低亮度行人轨迹和意图估计

doi:10.3969/j.issn.1671-7775.2022.04.005

摘要
图/表
参考文献(0)
相关文章 (1)

全文: PDF (16023 KB) HTML (1 KB)
输出: BibTeX | EndNote (RIS)

摘要夜晚行车环境比白天复杂，所以辅助驾驶系统对夜晚行车环境的感知尤为重要.对于行人识别而言，传统的行人轨迹和意图估计算法在夜间的适用性不足.为了能够在夜间估计行人的意图和轨迹，提出一种低亮度环境下近距离行人的意图和轨迹估计方法.使用新的图像增强算法以提高人脸特征识别率.对于增强后的图像，基于YOLOv3和Openpose提出匹配2类结果的算法.进一步采用卡尔曼滤波修正YOLOv3识别误差，再基于匹配结果提出一种行人面部朝向估计算法，然后在上述2点基础上提出估计行人意图新方法.结果表明：图像经过新增强算法处理后，提出的行人轨迹和意图估计方法对于低亮度环境有较好的适用性.

	服务

	把本文推荐给朋友
	加入我的书架
	加入引用管理器
	E-mail Alert
	RSS
	作者相关文章
	汪威
	罗石
	耿国庆
	刘军

关键词 ：近距离行人, 低亮度图像增强, 改进MSR算法, YOLOv3, Openpose

Abstract：The driving environment at night is more complex than that during the day, so the perception of the driving environment at night by the assistant driving system is very important. For pedestrian recognition, the traditional pedestrian trajectory and intention estimation algorithm is not suitable at night. To estimate pedestrian intention and trajectory at night, a method was proposed for estimating pedestrian intention and trajectory in short distance under low brightness environment. The new image enhancement algorithm was used to improve the face recognition rate. For the enhanced image, the matching algorithm was proposed based on YOLOv3 and Openpose. The Kalman filter was used to correct the YOLOv3 recognition error, and a pedestrian face orientation estimation algorithm was proposed based on the matching results to obtain a new method for estimating pedestrian intention. The results show that the proposed pedestrian trajectory and intention estimation method is suitable for low luminance environment after image enhancement.

Key words： low-distance pedestrian low brightness image enhancement improved MSR algorithm YOLOv3 Openpose

基金资助:国家自然科学基金资助项目（51675235）

作者简介: 汪威（1996—），男，江苏泰兴人，硕士研究生(1720583834@qq.com)，主要从事交通图像处理研究. 罗石（1969—），男，湖南衡山人，副教授(通信作者，luoshi@ujs.edu.cn)，主要从事汽车电子与控制技术研究.

引用本文:

汪威，罗石，耿国庆，刘军. 基于行人关键点的低亮度行人轨迹和意图估计[J]. 江苏大学学报（自然科学版）, 2022, 43(4): 400-406. WANG Wei， LUO Shi， GENG Guoqing， LIU Jun. Pedestrian trajectory and intention estimation under low brightness based on human key points[J]. Journal of Jiangsu University(Natural Science Eidtion) , 2022, 43(4): 400-406.

链接本文:

http://zzs.ujs.edu.cn/xbzkb/CN/10.3969/j.issn.1671-7775.2022.04.005 或 http://zzs.ujs.edu.cn/xbzkb/CN/Y2022/V43/I4/400

［1］	袁朝春,宋金行,何友国,等.基于行人轨迹预测的无人驾驶汽车主动避撞算法[J].江苏大学学报（自然科学版）,2021,42(1):1-8.
	YUAN C C, SONG J H, HE Y G, et al. Active collision avoidance algorithm of autonomous vehicle based on pedestrian trajectory prediction[J]. Journal of Jiangsu University（Natural Science Edition）,2021,42(1):1-8. (in Chinese)
［2］	SULISTIYO M D, KAWANISHI Y, DEGUCHI D, et al. ColAttnet: in reducing the ambiguity of pedestrian orientations on attributeaware semantic segmentation task[J]. IEEJ Transactions on Electrical and Electronic Engineering,2021,16(2):295-306.
［3］	CAO D, FU Y B. Using graph convolutional networks skeletonbased pedestrian intention estimation models for trajectory prediction[J].Journal of Physics:Conference Series，doi:10.1088/1742-6596/1621/1/012047.
［4］	冯欣,李永波,杨武.一种利用目标结构关系增强的行人重识别方法[J/OL].重庆理工大学学报(自然科学). http:∥kns.cnki.net/kcms/detail/50.1205.t.20220424.1455.002.html.
	FENG X, LI Y B, YANG W. A structural relationship enhanced person reidentification method\[J/OL\].Journal of Chongqing University of Technology (Natural Science). http:∥kns.cnki.net/kcms/detail/50.1205.t.20220424.1455.002.html. (in Chinese)
［5］	JOBSON D J, RAHMAN Z, WOODELL G A. A multiscale retinex for bridging the gap between color images and the human observation of scenes[J]. IEEE Transactions on Image Processing,1997,6(7):965-976.
［6］	刘富,刘璐,侯涛,等.基于优化MSR的夜间道路图像增强方法[J].吉林大学学报(工学版),2021,51(1):323-330.
	LIU F, LIU L, HOU T, et al. Night road image enhancement method based on optimized MSR[J]. Journal of Jilin University (Engineering and Technology Edition), 2021,51(1):323-330. (in Chinese)
［7］	LIN H N, SHI Z W. Multiscale retinex improvement for nighttime image enhancement[J].Optik,2014,125(24):7143-7148.
［8］	MA T T, YE W H, LENG S, et al. Solid waste surface feature enhancement method based on gamma correction and wavelet transform[J].Signal, Image and Video Processing,2021,15(7):1627-1634.
［9］	FENG L G, LIN L. Image denoising methods based on wavelet transform and threshold functions[J]. Journal of Multimimedia Processing Technologies, 2017, 8(1): 1-10.
［10］	朱铮涛,萧达安.基于非线性调整的伽马校正图像增强算法[J].计算机工程与设计,2018,39(9):2822-2826,2866.
	ZHU Z T, XIAO D A. Gammacorrected image enhancement algorithm based on nonlinear adjustment[J].Computer Engineering and Design, 2018,39(9):2822-2826,2866. (in Chinese)