雾天环境下前车车距测量方法研究

doi:10.3976/j.issn.1002-4026.20230034

山东科学 ›› 2024, Vol. 37 ›› Issue (1): 88-94.doi: 10.3976/j.issn.1002-4026.20230034

雾天环境下前车车距测量方法研究

盛雨婷()

合肥工业大学智能制造技术研究院,安徽合肥 230041

收稿日期:2023-02-22 出版日期:2024-02-20 发布日期:2024-01-26
作者简介:盛雨婷(1994—),女,硕士,初级工程师,研究方向为图像处理与自动控制。E-mail:623997509@qq.com

Study on the distance measurement of approaching vehicles in fog

SHENG Yuting()

Intelligent Manufacturing Institute of Hefei University of Technology, Hefei 230041, China

Received:2023-02-22 Online:2024-02-20 Published:2024-01-26

摘要/Abstract

摘要：

为解决雾天环境下道路上车辆与前车车距测量问题,构造车载雾天图像快速处理以及前车车距测量实验平台。以暗通道算法为基础,基于能见度图像分割算法估算大气光值,利用双边滤波细化折射率图,在分割区域上进行不同程度去雾,有效解决暗通道算法应用在道路图像上产生的色彩失真、对比度过低等问题。利用边缘检测算法、霍夫变换算法完成对车辆边框的检测,搭建测距模型测量出前方车辆的距离。结果表明,构造的平台能够在能见度小于100 m的浓雾环境下测量出前方车辆车距,并能及时告警。

关键词: 图像去雾, 能见度分割, 暗通道算法, 双边滤波, 边缘检测算法, 车距测量

Abstract:

To address the challenges related to distance measurement of an approaching vehicle in fog,we developed an experimental platform to rapid image processing and real-time distance measurement.Firstly,we down-sampled the images through the dark channel algorithm to estimate atmospheric light values. Then, we introduced a tolerance mechanism to deal with the bright regions that do not satisfy the dark channel prior. This tolerance mechanism corrected the estimate with incorrect refractive index of such regions and effectively mitigated the issues of color distortion and low contrast. Secondly, we detected the vertical edges of an approaching vehicle using the edge detection and the improved Hough transform algorithms. Finally, we measured the safe distance from the approaching vehicle using the model. The results shows that the platform developed in this study can effectively measure the distance of the approaching vehiclein fog with a visibility <100 m, and can alert drivers in a timely and effective manner.

Key words: haze removal, image down-sampling, dark channel algorithm, bilateral filtering, edge detection algorithm, vehicle distance measurement

中图分类号:

TP751

盛雨婷. 雾天环境下前车车距测量方法研究[J]. 山东科学, 2024, 37(1): 88-94.

SHENG Yuting. Study on the distance measurement of approaching vehicles in fog[J]. Shandong Science, 2024, 37(1): 88-94.

图/表 10

图1

图2

图3

图4

表1

图5

图6

图7

图8

表2

参考文献 13

[1]	潘晓东, 蒋宏, 高昂. 雾天高速公路交通事故成因分析及安全对策[J]. 交通标准化, 2006(10):200-203.DOI:10.3869/j.issn.1002-4786.2006.10.060.
[2]	胡平. 基于图像分割的交通图像快速去雾算法[J]. 计算机系统应用, 2014, 23(9): 134-138. DOI: 10.3969/j.issn.1003-3254.2014.09.025.
[3]	毛祥宇, 李为相, 丁雪梅. 基于天空分割的单幅图像去雾算法[J]. 计算机应用, 2017, 37(10):2916-2920. DOI: 10.11772/j.issn.1001-9081.2017.10.2916.
[4]	宋洪军, 陈阳舟, 郜园园. 基于车道线检测与图像拐点的道路能见度估计[J]. 计算机应用, 2012, 32(12):3397-3403. DOI: 10.3724/SP.J.1087.2012.03397.
[5]	TOPAL C, AKINLAR C. Edge Drawing: A combined real-time edge and segment detector[J]. Journal of Visual Communication and Image Representation, 2012, 23(6):862-872. DOI: 10.1016/j.jvcir.2012.05.004.
[6]	AKINLARC, TOPAL C. EDLines: a real-time line segment detector with a false detection control[J]. Pattern Recognition Letters, 2011, 32(13):1633-1642. DOI: 10.1016/j.patrec.2011.06.001.
[7]	KOHLER R. A segmentation system based on thresholding[J]. Computer Graphics and Image Processing, 1981, 15(2):319-338.DOI:10.1016/S0146-664X(81)80015-9.
[8]	李炎炎, 龙伟, 覃宏超, 等. 基于车行视程与大气光快速估值的车载视频去雾算法[J]. 工程科学与技术, 2017, 49(3):217-222. DOI: 10.15961/j.jsuese.201601407.
[9]	陆士猛, 刘昌锦. 无人机侦察图像快速去雾算法[J]. 红外技术, 2015, 37(10):847-851.DOI: 10.11846/j.issn.1001_8891.201510008.
[10]	HE K, SUN J, TANG X. Single image haze removal using dark channel prior[J]. Pattern Analysis and Machine Intelligence, IEEE Transactions on, 2011, 33(12): 2341-2353.DOI:10.1109/TPAMI.2010.168.
[11]	郭璠, 蔡自兴. 图像去雾算法清晰化效果客观评价方法[J]. 自动化学报, 2012, 38(9):1410-1419.DOI: 10.3724/SP.J.1004.2012.01410.
[12]	CHENR B, ZHAO D, WANG Q L. Research on the algorithm of the front vehicle speed based on vehicle video[C]// 29th Chinese Control and Decision Conference (CCDC). Chongqing, China: IEEE, 2017:7719-7724. DOI: 10.1109/CCDC.2017.7978591.
[13]	佟卓远. 基于机器视觉的前方车辆检测与测距系统设计[D]. 哈尔滨: 哈尔滨工业大学, 2015.

算法	复原后的新增可见之比	可见边的规范化梯度均值	方差	运算时间/s
暗通道算法	0.58	2.20	43.06	28.238
本文算法	1.57	2.21	41.02	0.051

项目	薄雾	中雾	浓雾
原图测距/m	13.93	46.88	—
原图像素差	62.00	16.00	—
去雾后测距/m	14.01	18.29	9.49
去雾后像素差	61.50	41.00	79.00
测距误差/%	0.006	0.610	—

雾天环境下前车车距测量方法研究

Study on the distance measurement of approaching vehicles in fog

RichHTML

PDF (PC)

赞

可视化

摘要/Abstract

引用本文

使用本文

图/表 10

参考文献 13

相关文章 1

Metrics

本文评价

推荐阅读 0