Optics and Precision Engineering, Volume. 32, Issue 5, 752(2024)
Binocular vision SLAM with fused point and line features in weak texture environment
Figures & Tables(11)
Fig. 1. Framework of binocular vision odometer system integrating point and line features
Fig. 3. Schematic diagram of binocular line feature triangulation method based on geometric constraints
Fig. 4. Schematic diagram of weak texture dataset collection devices
Fig. 6. Comparison of binocular matching algorithm under weak texture data
Fig. 7. Comparison of image construction and positioning effects between improved binocular point line fusion visual SLAM and ORBSLAM2 algorithms
Table 1. [in Chinese]
View tableView in ArticleTable 1. [in Chinese]
算法1 ：基于改进LSD的线特征提取算法输入：图像I，阈值T
输出：线特征集合
1. 对输入图像应用高斯平滑。
2. 在图像的每个像素处计算梯度幅度和方向。
3. 利用区域生长法得到线矩形区域
4. 遍历图像中的剩余像素：
a. 使用梯度方向计算局部梯度方向。
b. 计算最接近局部梯度方向的直线方向。
c. 将像素分配给与最接近方向匹配的直线。
5. 输出线矩形区域的NFA<ε的线段。
6. 遍历图像中的每个直线段：
a. 计算直线段长度。
b. 如果直线段长度小于阈值T：
c. 如果直线段满足
式（3） ，则与相邻的直线段合并继续循环；d. 如果直线段不与任何其他直线段平行或垂直，则丢弃它。
7. 返回剩余的直线段。
Table 1. Comparison between proposed method and line segment matching method based on LBD
View tableView in ArticleTable 1. Comparison between proposed method and line segment matching method based on LBD
Dataset LSD+LBD FLD+LBD Ours（无邻域约束） Ours（无区域约束） Ours 数量 精度/% 数量 精度/% 数量 精度/% 精度/% 精度/% KITTI/seq 00 86 66 59 51 45 72 78 81 KITTI/seq 01 91 71 67 56 56 70 85 87 KITTI/seq 02 92 62 71 49 62 77 80 82 KITTI/seq 03 88 63 73 59 58 72 79 82 KITTI/seq 04 79 69 64 45 63 73 77 79 Euroc/VR 01 80 70 58 51 51 86 90 91 Euroc/VR 02 83 83 51 66 44 88 91 93 EuRoC/MH 01 93 78 71 60 57 78 83 85 EuRoC/MH 02 99 71 79 61 68 72 80 82 自建弱纹理数据集 45 72 39 56 32 87 90 91 平均值 84 71 63 55 54 79 83 85
Table 2. Comparison of time consumption of different algorithms
View tableView in ArticleTable 2. Comparison of time consumption of different algorithms
方 法 线特征提取 线特征匹配 总时间 LSD+LBD 34.2 17.5 51.7 FLD+LBD 18.5 15.3 33.8 Ours 33.1 7.4 40.5
Tools
Get Citation
Copy Citation Text
Kun GONG, Xin XU, Xiaoqing CHEN, Yuelei XU, Zhaoxiang ZHANG. Binocular vision SLAM with fused point and line features in weak texture environment[J]. Optics and Precision Engineering, 2024, 32(5): 752
Paper Information
Category:
Received: Apr. 21, 2023
Accepted: --
Published Online: Apr. 2, 2024
The Author Email: Yuelei XU (xuyuelei@nwpu.edu.cn)
© Copyright 2018-2021 | Chinese Laser Press.
All Rights Reserved 沪ICP备15018463号-20