用于智能驾驶的动态场景视觉显著性多特征建模方法

doi:10.11947/j.AGCS.2021.20210266

测绘学报 ›› 2021, Vol. 50 ›› Issue (11): 1500-1511.doi: 10.11947/j.AGCS.2021.20210266

用于智能驾驶的动态场景视觉显著性多特征建模方法

詹智成^1,2, 董卫华¹

1. 北京师范大学地理科学学部, 北京 100875;
2. 根特大学地理学院, 比利时根特 9000

收稿日期:2021-05-13 修回日期:2021-09-28 发布日期:2021-12-07
通讯作者: 董卫华 E-mail:dongweihua@bnu.edu.cn
作者简介:詹智成(1994—),男,博士生,研究方向为地理信息系统和地理空间认知。
基金资助:
国家自然科学基金（41871366）；国家留学基金委项目资助（201906040236）；地理信息工程国家重点实验室、自然资源部测绘科学与地球空间信息技术重点实验室联合资助基金（2021-04-03）

A multi-feature approach for modeling visual saliency of dynamic scene for intelligent driving

ZHAN Zhicheng^1,2, DONG Weihua¹

1. Faculty of Geographical Science, Beijing Normal University, Beijing 100875, China;
2. Department of Geography, Ghent University, Ghent 9000, Belgium

Received:2021-05-13 Revised:2021-09-28 Published:2021-12-07
Supported by:
The National Natural Science Foundation of China (No. 41871366);The China Scholarship Council (No. 201906040236);The State Key Laboratory of Geographic Information Engineering and the Key Laboratory of Surveying and Mapping Science and Geospatial Information Technology of the Ministry of Natural Resources Jointly Funded Project (No. 2021-04-03)

摘要/Abstract

摘要： 驾驶场景的视觉显著性建模是智能驾驶的重要研究方向。现有的静态和虚拟场景的视觉显著性建模方法不能适应真实驾驶环境下道路场景实时性、动态性和任务驱动特性。构建真实驾驶环境的动态场景视觉显著性模型是目前研究的挑战。从驾驶环境的特点与驾驶员的视觉认知规律出发，本文提取道路场景的低级视觉特征、高级视觉特征和动态视觉特征，并结合速度和道路曲率两个重要影响因素，建立了多特征逻辑回归模型（logistic regression，LR）计算驾驶场景视觉显著性。使用AUC值对模型进行评价，结果显示精度达到了90.43%，与传统的算法相比具有明显的优势。

关键词: 视觉显著性, 驾驶场景, 驾驶环境, 动态性

Abstract: Visual saliency modeling of driving scenes is an important research direction in intelligent driving, especially in the areas of assisted driving and automatic driving. The existing visual saliency modeling methods for static and virtual scenes cannot adapt to the real-time, dynamic and task-driven characteristics of road scenes in real driving environments. Building a visual saliency model of dynamic road scenes in real driving environments is a challenge for current research. Starting from the characteristics of driving environment and driver’s visual cognitive law, this paper extracts low-level visual features, high-level visual features and dynamic visual features of road scenes, and combines two influencing factors of speed and road curvature to build a visual saliency calculation model of driving scenes based on logistic regression model (LR). In this paper, the AUC value is used to evaluate the model, and the results showed an accuracy of 90.43%, which is significant advantage over traditional algorithms.

Key words: visual saliency, driving scene, driving environment, dynamics

中图分类号:

P208

詹智成, 董卫华. 用于智能驾驶的动态场景视觉显著性多特征建模方法[J]. 测绘学报, 2021, 50(11): 1500-1511.

ZHAN Zhicheng, DONG Weihua. A multi-feature approach for modeling visual saliency of dynamic scene for intelligent driving[J]. Acta Geodaetica et Cartographica Sinica, 2021, 50(11): 1500-1511.

参考文献

[1] BORJI A. Boosting bottom-up and top-down visual features for saliency estimation[C]//Proceedings of 2012 IEEE Conference on Computer Vision and Pattern Recognition. Providence, RI, USA: IEEE, 2012: 438-445.
[2] NOTHDURFT H C. Salience of feature contrast[M]//Neurobiology of Attention. Amsterdam: Elsevier, 2005: 233-239.
[3] HENDERSON J M, HOLLINGWORTH A. High-level scene perception[J]. Annual Review of Psychology, 1999, 50(1): 243-271.
[4] CADUFF D, TIMPF S. On the assessment of landmark salience for human navigation[J]. Cognitive Processing, 2008, 9(4): 249-267.
[5] LIAO Hua, DONG Weihua, PENG Chen, et al. Exploring differences of visual attention in pedestrian navigation when using 2D maps and 3D geo-browsers[J]. Cartography and Geographic Information Science, 2017, 44(6): 474-490.
[6] LIAO Hua, DONG Weihua. An exploratory study investigating gender effects on using 3D maps for spatial orientation in wayfinding[J]. ISPRS International Journal of Geo-Information, 2017, 6(3): 60.
[7] OHM C, MVLLER M, LUDWIG B, et al. Where is the Landmark? Eye tracking studies in large-scale indoor environments[C]//Proceedings of the 2nd International Workshop on Eye Tracking for Spatial Research (in conjunction with GIScience 2014). Vienna, Austria: CEUR Workshop Proceedings, 2014: 47-51.
[8] LIAO Hua, DONG Weihua, HUANG Haosheng, et al. Inferring user tasks in pedestrian navigation from eye movement data in real-world environments[J]. International Journal of Geographical Information Science, 2019, 33(4): 739-763.
[9] WANG Shuihua, TIAN Yingli. Indoor signage detection based on saliency map and bipartite graph matching[C]//Proceedings of 2011 IEEE International Conference on Bioinformatics and Biomedicine Workshops(BIBMW). Atlanta, GA, USA: IEEE, 2011: 518-525.
[10] DAI Yuchao, ZHANG Jing, HE Mingyi, PORIKLI Fatih, LIU Bowen. Salient Object Detection from Multi-spectral Remote Sensing Images with Deep Residual Network[J]. Journal of Geodesy and Geoinformation Science, 2019, 2(2): 101-110.
[11] BROUWER W H, WATERINK W, VAN WOLFFELAAR P C, et al. Divided attention in experienced young and older drivers: lane tracking and visual analysis in a dynamic driving simulator[J]. Human Factors, 1991, 33(5): 573-582.
[12] VIDAL R, RAVICHANDRAN A. Optical flow estimation & segmentation of multiple moving dynamic textures[C]//Proceedings of 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR’05). San Diego, CA, USA: IEEE, 2005: 516-521.
[13] JENSEN S S. Driving patterns and emissions from different types of roads[J]. Science of the Total Environment, 1995, 169(1/2/3): 123-128.
[14] LIU Jingnan, ZHAN Jiao, GUO Chi, et al. Data logic structure and key technologies on intelligent high-precision map[J]. Journal of Geodesy and Geoinformation Science, 2020,3(3):1-17.
[15] DOSHI A, TRIVEDI M. Investigating the relationships between gaze patterns, dynamic vehicle surround analysis, and driver intentions[C]//Proceedings of 2009 IEEE Intelligent Vehicles Symposium. Xi’an, China: IEEE, 2009: 887-892.
[16] PALINKO O, KUN A L, SHYROKOV A, et al. Estimating cognitive load using remote eye tracking in a driving simulator[C]//Proceedings of 2010 Symposium on Eye-Tracking Research & Applications. Austin, Texas, NY, USA: ACM Press, 2010: 141-144.
[17] RECARTE M A, NUNES L M. Effects of verbal and spatial-imagery tasks on eye fixations while driving[J]. Journal of Experimental Psychology Applied, 2000, 6(1): 31-43.
[18] PALAZZI A, ABATI D, CALDERARA S, et al. Predicting the driver’s focus of attention: the DR(eye)VE project[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2019, 41(7): 1720-1733.
[19] 吴晓峰. 基于驾驶员工作负荷的公路线形安全性评价[D]. 西安: 长安大学, 2009. WU Xiaofeng. Highway geometry safety evaluation based on driver workload[D]. Xi’an: Changan University, 2009.
[20] WERNEKE J, VOLLRATH M. What does the driver look at? The influence of intersection characteristics on attention allocation and driving behavior[J]. Accident Analysis & Prevention, 2012, 45: 610-619.
[21] DENG Tao, YANG Kaifu, LI Yongjie, et al. Where does the driver look? top-down-based saliency detection in a traffic driving environment[J]. IEEE Transactions on Intelligent Transportation Systems, 2016, 17(7): 2051-2062.
[22] SHASHUA A, GDALYAHU Y, HAYUN G. Pedestrian detection for driving assistance systems: single-frame classification and system level performance[C]//Proceedings of 2004 IEEE Intelligent Vehicles Symposium. Parma, Italy: IEEE, 2004: 1-6.
[23] ZHONG Zhun, LEI Mingyi, CAO Donglin, et al. Class-specific object proposals re-ranking for object detection in automatic driving[J]. Neurocomputing, 2017, 242: 187-194.
[24] MILANÉS V, LLORCA D F, VILLAGRÁ J, et al. Intelligent automatic overtaking system using vision for vehicle detection[J]. Expert Systems With Applications, 2012, 39(3): 3362-3373.
[25] RISACK R, KLAUSMANN P, KRVGER W, et al. Robust lane recognition embedded in a real-time driver assistance system[C]//Proceedings of 1998 IEEE International Conference on Intelligent Vehicles. Piscataway, NJ, USA: IEEE,1998:35-40.
[26] JEONG S G, KIM C S, LEE D Y, et al. Real-time lane detection for autonomous vehicle[C]//Proceedings of 2001 IEEE International Symposium on Industrial Electronics. Pusan, Korea (South): IEEE, 2001: 1466-1471.
[27] PAULO C F, CORREIA P L. Automatic detection and classification of traffic signs[C]//Proceedings of the 8th International Workshop on Image Analysis for Multimedia Interactive Services. Santorini, Greece: IEEE, 2007: 11.
[28] LOBO J M, JIMÉNEZ-VALVERDE A, REAL R. AUC: a misleading measure of the performance of predictive distribution models[J]. Global Ecology and Biogeography, 2008, 17(2): 145-151.
[29] DENG Tao, YAN Hongmei, LI Yongjie. Learning to boost bottom-up fixation prediction in driving environments via random forest[J]. IEEE Transactions on Intelligent Transportation Systems, 2018, 19(9): 3059-3067.
[30] PETERS R J, IYER A, ITTI L, et al. Components of bottom-up gaze allocation in natural images[J]. Vision Research, 2005, 45(18): 2397-2416.
[31] HORREY W J, WICKENS C D, CONSALUS K P. Modeling drivers’ visual attention allocation while interacting with in-vehicle technologies[J]. Journal of Experimental Psychology: Applied, 2006, 12(2): 67-78.
[32] TAWARI A, KANG B. A computational framework for driver’s visual attention using a fully convolutional architecture[C]//Proceedings of 2017 IEEE Intelligent Vehicles Symposium (IV). 2017, Los Angeles, CA, USA: IEEE, 2017: 887-894.
[33] XIE Yuan, LIU Lifeng, LI Cuihua, et al. Unifying visual saliency with HOG feature learning for traffic sign detection[C]//Proceedings of 2009 IEEE Intelligent Vehicles Symposium. Xi’an, China: IEEE, 2009: 24-29.
[34] MOHANDOSS T, PAL S, MITRA P. Visual attention for behavioral cloning in autonomous driving[C]//Proceedings of 11th International Conference on Machine Vision (ICMV 2018). Munich,Germany: SPIE, 2019: 361-371.
[35] HORREY W J, WICKENS C D, CONSALUS K P. Modeling drivers’ visual attention allocation while interacting with in-vehicle technologies[J]. Journal of Experimental Psychology: Applied, 2006, 12(2): 67-78.
[36] EYRAUD R, ZIBETTI E, BACCINO T. Allocation of visual attention while driving with simulated augmented reality[J]. Transportation Research Part F: Traffic Psychology and Behaviour, 2015, 32: 46-55.
[37] 毛征宇,刘中坚.一种三次均匀B样条曲线的轨迹规划方法[J]. 中国机械工程, 2010, 21(21): 2569-2572,2577. MAO Zhengyu, LIU Zhongjian. A trajectory planning method for cubic uniform B-spline curve[J]. China Mechanical Engineering, 2010, 21(21): 2569-2572, 2577.
[38] 李泳波. 基于RANSAC的道路消失点自适应检测算法[J]. 中国科技信息, 2017(13): 80-82, 13. LI Yongbo. An adaptive detection algorithm for road vanishing points based on RANSAC [J]. China Science and Technology Information. 2017(13):80-82,13.
[39] MOGHADAM P, STARZYK J A, WIJESOMA W S. Fast vanishing-point detection in unstructured environments[J]. IEEE Transactions on Image Processing, 2012, 21(1): 425-430.
[40] RASMUSSEN C. Grouping dominant orientaions for ill-structured road following[C]//Proceedings of 2004 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Washington, DC, USA:IEEE, 2004:I.
[41] KONG Hui, AUDIBERT J Y, PONCE J. Vanishing point detection for road detection[C]//Proceedings of 2009 IEEE Conference on Computer Vision and Pattern Recognition. Miami, FL, USA: IEEE, 2009: 96-103.
[42] CHEN L C, PAPANDREOU G, KOKKINOS I, et al. DeepLab: semantic image segmentation with deep convolutional nets, atrous convolution, and fully connected CRFs[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2018, 40(4): 834-848.
[43] CORDTS M, OMRAN M, RAMOS S, et al. The cityscapes dataset for semantic urban scene understanding[C]//Proceedings of 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Las Vegas, NV, USA: IEEE, 2016: 3213-3223.
[44] MAULUD D, ABDULAZEEZ A M. A review on linear regression comprehensive in machine learning[J]. Journal of Applied Science and Technology Trends, 2020,1(4): 140-147.

用于智能驾驶的动态场景视觉显著性多特征建模方法

A multi-feature approach for modeling visual saliency of dynamic scene for intelligent driving

RichHTML

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 2

编辑推荐

Metrics

本文评价

[1]	陈一祥, 秦昆, 张晔, 袁媛. 视觉显著性与知觉组织相结合的高分影像居民地提取方法[J]. 测绘学报, 2017, 46(12): 1959-1968.
[2]	沈小乐邵振峰田英洁. 纹理特征与视觉注意相结合的建筑区提取[J]. 测绘学报, 2014, 43(8): 842-847.