|
|
Detecting obstacles in vegetation by multi spectral fusion |
WANG Sheng,XIANG Zhi yu |
Department of Information Science and Electronic Engineer, Zhejiang University, Hangzhou 310027, China |
|
|
Abstract A detection method that integrates distribution features of threedimensional point cloud and multispectral signature was proposed, for the obstacle detection in vegetationcovered environment. A multisensor system consisting of color camera, infrared camera and threedimensional LIDAR was built. The integration of threedimensional point cloud data and image pixels was achieved,through a joint calibration method by camera and threedimensional LIDAR. A new spectral signature of IRcolor joint channel was proposed and used to classify vegetation and nonvegetation objects together with Gaussian Mixture Model,based on multispectrum data analysis and normalized difference vegetation index (NDVI). Experimental results showed that the changes of lighting conditions and the interference of ground points had a significant impact on the results. the detection effect was significantly improved, by adding normalized light intensity of infrared light and weighted feature information, Experiments were conducted in several typical scenarios. Results showed the detection effect by the method was better than the one by NDVI.
|
Published: 01 November 2015
|
|
基于多谱融合的植被环境中障碍物检测
针对植被环境下的障碍物检测,提出联合三维点云分布特征和多光谱特征的检测方法.构建彩色相机、红外相机和三维激光雷达构成的多传感器系统.通过相机与三维激光雷达联合标定的方法,实现三维点云数据与图像像素信息的融合.基于多光谱数据分析,在归一化植被差分指数(NDVI)的基础上,提出新的红外彩色通道联合光谱特征,结合混合高斯模型对植被和非植被进行分类.在实验中发现光照条件的变化和地面点的干扰对结果有很大影响.通过加入红外光强归一化和特征信息加权之后检测效果得到了明显改善.在多个典型的场景中进行实验,结果表明,检测效果比基于NDVI的方法好.
|
|
[1] MACEDO J, MANDUCHI R, MATTHIES L. Ladarbased discrimination of grass from obstacles for autonomous navigation[C]∥ ISER 2000, Hawaii, USA:Springer Berlin Heidelberg, 2000, 271: 111-120.
[2] MANDUCHI R, CASTANO A, TALUKDER A, et al. Obstacle detection and terrain classification for autonomous offroad navigation[J]. Autonomous Robots, 2005, 18(1): 81-102.
[3] HEBERT M, VANDAPEL N. Terrain classification techniques from ladar data for autonomous navigation[C]∥ Collaborative Technology Alliance Workshop. Adelphi, Maryland, USA: Army Research Laboratory, 2003: 4-11.
[4] VANDAPEL N, HUBER D F, KAPURIA A, et al. Natural terrain classification using 3d ladar data[C]∥ Proc. of IEEE International Conference on Robotics and Automation (ICRA’04).New Orleans, LA, USA: IEEE Press, 2004: 5117-5122.
[5] RICHARDSON A J, WIEGAND C L. Distinguishing vegetation from soil background information [J]. Photogrammetric Engineering and Remote Sensing, 1977, 43(12):15411552.
[6] HUETE A R. A soil adjusted vegetation index (SAVI) [J]. Remote Sensing of Environment, 1988, 25: 295-309.
[7] DIMA C S, VANDAPEL N, HEBERT M. Classifier fusion for outdoor obstacle detection[C]∥ Proc. of IEEE International Conference on Robotics and Automation (ICRA’04). New Orleans, LA, USA: IEEE Press, 2004, 1: 665-671.
[8] BRADLEY D, THAYER S, STENTZ A T, et al. Vegetation detection for mobile robot navigation[R]. Pittsburgh, USA: Carnegie Mellon University, Robotics Institute, February 2004.
[9] BRADLEY D M, UNNIKRISHNAN R, BAGNELL J. Vegetation detection for driving in complex environments[C]∥ Proc. of IEEE International Conference on Robotics and Automation (ICRA’07). Roma, Italy: IEEE Press, 2007: 503-508.
[10] NGUYEN D V, KUHNERT L, SCHLEMPER J, et al. Terrain classification based on structure for autonomous navigation in complex environments[C]∥ Proc. of the 3rd International Conference on Communications and Electronics (ICCE). Nha Trang: IEEE Press, 2010: 163-168.
[11] NGUYEN D V, KUHNERT L, JIANG T, et al. Vegetation detection for outdoor automobile guidance[C]∥ Proc. of International Conference on Industrial Technology (ICIT). Auburn, AL: IEEE Press, 2011: 358-364.
[12] UNNIKRISHNAN R, HEBERT M. Fast Extrinsic Calibration of a Laser Rangefinder to a Camera[R]. Pittsburgh, USA: Carnegie Mellon University, Robotics Institute, July 2005.
[13] SHULL C A. A spectrophotometric study of the reflection of light from leaf surfaces[J]. Botanical Gazette, 1929, 87(5): 583-607.
[14] CLARK R N, SWAYZE G A, LIVO K E, et al. Imaging spectroscopy: Earth and planetary remote sensing with the USGS Tetracorder and expert systems[J]. Journal of Geophysical Research: Planets, 2003, 108(12): 5131.
[15] UNSALAN C, BOYER K L, Linearized vegetation indices based on a formal statistical framework[J]. IEEE Trans. on Geoscience and Remote Sensing, 2004, 42(7): 1575-1585.
[16] KREIGLER F J, MALILA W A, NALEPKA R F, et al. Preprocessing transformations and their effects on multispectral recognition [C]∥ Proc. of the Sixth International Symposium on Remote Sensing of Environment. Ann Arbor, Michigan, USA, Environmental Research Institute of Michigan, 1969: 97-131.
[17] BILMES J A. A gentle tutorial of the EM algorithm and its application to parameter estimation for Gaussian mixture and hidden Markov models[R]. California, USA: University of Berkeley, International Computer Science Institute, 1997.
[18] NAVARROSERMENT L E, MERTZ C, HEBERT M. Pedestrian detection and tracking using threedimensional ladar data[J]. The International Journal of Robotics Research, 2010, 29(12): 1516-1528. |
|
Viewed |
|
|
|
Full text
|
|
|
|
|
Abstract
|
|
|
|
|
Cited |
|
|
|
|
|
Shared |
|
|
|
|
|
Discussed |
|
|
|
|