|
|
论文:2022,Vol:40,Issue(6):1431-1439 |
|
|
引用本文： |
|
|
陈星, 张文海, 杨林, 郑讯佳. 结合立体视觉的非结构化道路消失点检测研究[J]. 西北工业大学学报 |
|
|
CHEN Xing, ZHANG Wenhai, YANG Lin, ZHENG Xunjia. Research on vanishing point detection of unstructured road scene combined with stereo vision[J]. Journal of Northwestern Polytechnical University |
|
|
|
|
|
|
|
| 结合立体视觉的非结构化道路消失点检测研究 |
|
| 陈星1,2, 张文海2, 杨林3, 郑讯佳1 |
|
1. 重庆文理学院 智能制造工程学院, 重庆 402160;
2. 重庆交通大学 机电与汽车工程学院, 重庆 400074;
3. 重庆长安工业(集团)有限责任公司 特种车辆研究所, 重庆 400023 |
| 摘要: |
| 消失点检测是基于视觉的无人车辆自主导航的重要组成部分。由于非结构化场景存在缺乏清晰的道路线和复杂的背景干扰等问题,现有检测方法普遍存在精度低、计算时间长的缺点。因此,针对非结构化道路特点,提出了一种结合立体视觉的消失点检测方法。采用双目立体视觉技术获得道路图像的视差图,使用广度优先算法快速估计出道路图像的背景区域;设计四方向五尺度的Gabor滤波器组估计像素响应幅值,并通过幅值校正减少检测误差;结合背景区域设计一系列投票点选择策略,来剔除背景区域的干扰,提高算法精度;采用动态调整候选点范围策略,减少消失点的搜索范围,从而提高算法效率;设计了一种角度优先投票函数,将在投票空间中获得最大票数的候选点视为消失点。结果表明,改进的方法在复杂背景干扰的场景下具有较好的鲁棒性,在检测速度和检测精度上都有显著提升。 |
| 关键词:
消失点检测
立体视觉
广度优先
Gabor滤波器组
动态调整
背景干扰
|
|
| Research on vanishing point detection of unstructured road scene combined with stereo vision |
|
| CHEN Xing1,2, ZHANG Wenhai2, YANG Lin3, ZHENG Xunjia1 |
|
1. School of Intelligent Manufacturing, Chongqing University of Arts and Sciences, Chongqing 402160, China;
2. School of Mechatronics and Vehicle Engineering, Chongqing Jiaotong University, Chongqing 400074, China;
3. Special Vehicle Research Institute, Chongqing Changan Industry (Group) Co., Ltd, Chongqing 400023, China |
| Abstract: |
| Vanishing point detection is an important part of visual-based autonomous navigation of unmanned vehicles. Due to the lack of clear routes and complex background interference in unstructured road scenes, existing detection methods generally have the shortcomings of low accuracy and long calculation time. Therefore, according to the characteristics of unstructured road, a vanishing point detection method combined with stereo vision is proposed in this paper. Binocular stereo vision technology is used to obtain the parallax image of road image, and breadth-first algorithm is applied to the parallax image to estimate the background area of road image quickly. Gabor filter banks with 4 directions and 5 scales are designed to estimate the pixel response amplitude, and the detection error is reduced by amplitude correction. Combined with the background region, a series of voting place selection strategies are designed to eliminate the interference of the background region and improve the accuracy of the algorithm. A strategy of dynamically adjusting the range of candidate points is proposed to reduce the search range of vanishing points to improve the efficiency of the algorithm. Finally, an angle preference voting function is designed, which considers the candidate with the largest number of votes in the voting space as the vanishing point. Simulation results show that the improved method has better robustness in the scene with complex background interference, and significantly improves the detection speed and accuracy of vanishing point detection. |
| Key words:
vanishing point
stereo vision
breadth-first
Gabor filter banks
dynamic adjustment
background interference
|
|
| 收稿日期: 2021-12-29
修回日期:
|
| DOI: 10.1051/jnwpu/20224061431 |
| 基金项目: 重庆市教委科学技术研究项目(KJZD-K202101301)、中国博士后科学基金(2019T120813,2018M643420)与重庆文理学院人才引进项目(R2021SZZ01)资助 |
| 通讯作者: 张文海(1995—),重庆文理学院硕士研究生,主要从事图像处理及通路识别研究。e-mail:1064104787@qq.com
Email：1064104787@qq.com |
| 作者简介: 陈星(1985—),重庆文理学院副教授、硕士生导师,主要从事智能车辆自动驾驶技术与理论研究
|
|
| 相关功能 |
|
|
|
|
| 作者相关文章 |
|
| 陈星 在本刊中的所有文章 |
| 张文海 在本刊中的所有文章 |
| 杨林 在本刊中的所有文章 |
| 郑讯佳 在本刊中的所有文章 |
|
|
|
|
|
|
|
|
参考文献: |
|
|
[1] 王永忠, 文成林. 基于Haar纹理的非结构化道路消失点检测[J]. 中国图象图形学报, 2013, 18(4):27-36 WANG Yongzhong, WEN Chenglin. Unstructured road vanishing point detection based on Haar texture[J]. Chinese Journal of Image and Graphics, 2013, 18(4):27-36 (in Chinese)
[2] WANG Y, TEOHE K, SHEN D. Lane detection and tracking using B-snake[J]. Image and Vision Computing, 2004, 22(4):269-280
[3] NIETO M, SALGADO L. Real time vanishing point estimation in road sequences using adaptive steerable filter banks[C]//International Conference on Advanced Concepts for Intelligent Vision Systems, 2007
[4] 韩冰心, 卢惠民, 于清华, 等. 曼哈顿世界环境下消失点非线性优化估计算法[J]. 中国图图形学报, 2021, 26(12):2931-2940 HAN Bingxin, LU Huimin, YU Qinghua, et al. Nonlinear optimal vanishing point estimation algorithm in Manhattan world[J]. Chinese Journal of Image and Graphics, 2021,26(12):2931-2940 (in Chinese)
[5] RASMUSSEN C. Grouping dominant orientations for structured road following[C]//Proceedings of IEEE Conference on Computer Vision and Patten Recognition, 2004
[6] KONG H, AUDIBRT J Y. General road detection from a single image[J]. IEEE Trans on Image Processi, 2010, 19(8):2211-2220
[7] MOGHADA P, STARZYK J A, WIJESOMA W S. Fast vanishing point detection in unstructured environments[J]. IEEE Trans on Image Process, 2012, 21(1):425-430
[8] JIN J S, JIN X W, FANG F F. Fast and robust vanishing point detection for unstructured road following[J]. IEEE Trans on Intelligent Transportation Systems, 2016, 17(4):970-979
[9] 黄俊, 侯北平, 董霏, 等. 基于方向纹理的非结构化道路消失点检测研究[J]. 图学学报,2019,40(1):131-136 HUANG Jun, HOU Beiping, DONG Fei, et al. Research on vanishing point detection of unstructured road based on directional texture[J]. Journal of Graphics, 2019, 40(1):131-136 (in Chinese)
[10] 王先杰. 基于消失点的非结构化道路检测算法研究[D]. 淮南:安徽理工大学,2020 WANG Xianjie. Research on non-squiration road detection algorithm based on vain point[D]. Huainan:Anhui University of Technology, 2020 (in Chinese)
[11] 俞骏威, 张黎明, 陈凯, 等. 基于道路消失点的远距离路面微小障碍物检测[J]. 同济大学学报, 2020, 47(1):213-216 YU Junwei, ZHANG Liming, CHEN Kai, et al. Remote detection of small obstacles on road surface based on road vanishing point[J]. Journal of Tongji University, 2020, 47(1):213-216 (in Chinese)
[12] LIU Yinbo, ZENG Ming, MENG Qinghao. Unstructured road vanishing point detection using convolutional neural networks and heatmap regression[J]. IEEE Trans on Instrumentation and Measurement, 2021, 70(2):91-99
[13] 张欢, 安利, 张强, 等. SGBM算法BM算法分析研究[J]. 测绘与空间地理信息, 2016, 39(10):214-216 ZHANG Huang, AN Li, ZHANG Qiang, et al. SGBM algorithm and BM algorithm analysis and research[J]. Geomatics and Spatial Information Technology, 2016, 39(10):214-216 (in Chinese)
[14] RASMUSSEN C. Texture-based vanishing point voting for road shape estimation[C]//British Machine Vision Conference, 2009
[15] KONG H, AUIDIBERT J Y, PONCE J. Vanishing point detection for road detection[C]//IEEE Conference on Computer Vision and Pattern Recognition, 2009
[16] TRUNG H B, EITAKU N, TAKESHI S. A texture-based local soft voting method for vanishing point detection from a single road image[J]. IEICE Trans on Information and Systems, 2013, 96:690-698
[17] MENZE M, GEIGER A. Object scene flow for autonomous vehicles[C]//IEEE Conference on Computer Vision and Pattern Recognition, Piscataway, 2015
[18] DING W, LI Y, LIU H. Efficient vanishing point detection method in unstructured road environments based on dark channel prior[J]. IET Computer Vision, 2016, 10(8):852-860 |
|
|
|
|
|
|
|
