期刊
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
卷 23, 期 7, 页码 7841-7854出版社
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/TITS.2021.3073151
关键词
Intelligent Vehicles; ground segmentation; coarse-to-fine MRF; graph cut; real-time
资金
- National Key Research and Development Program of China [2020AAA0108103, 2016YFD0701401, 2017YFD0700303, 2018YFD0700602]
- Youth Innovation Promotion Association of the Chinese Academy of Sciences [2017488]
- Key Supported Project in the 13th Five-Year Plan of Hefei Institutes of Physical Science, Chinese Academy of Sciences [KP-2019-16]
- Natural Science Foundation of Anhui Province [1508085MF133]
- Technological Innovation Project for New Energy and Intelligent Networked Automobile Industry of Anhui Province
This paper proposes a fast point cloud ground segmentation approach based on a coarse-to-fine Markov random field (MRF) method, which utilizes an improved local feature extraction algorithm to achieve more accurate and computationally efficient ground segmentation. Experimental results demonstrate that the proposed approach outperforms traditional methods and is faster compared to graph-based methods.
Ground segmentation is an important preprocessing task for autonomous vehicles (AVs) with 3D LiDARs. However, the existing ground segmentation methods are very difficult to balance accuracy and computational complexity. This paper proposes a fast point cloud ground segmentation approach based on a coarse-to-fine Markov random field (MRF) method. The method uses the coarse segmentation result of an improved local feature extraction algorithm instead of prior knowledge to initialize an MRF model. It provides an initial value for the fine segmentation and dramatically reduces the computational complexity. The graph cut method is then used to minimize the proposed model to achieve fine segmentation. Experiments on two public datasets and field tests show that our approach is more accurate than both methods based on features and MRF and faster than graph-based methods. It can process Velodyne HDL-64E data frames in real-time (24.86 ms, on average) with only one thread of the 17-8700 CPU. Compared with methods based on deep learning, it has better environmental adaptability.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据