Terrain Classification of LiDAR Point Cloud Based on Multi-Scale Features and PointNet

Zhongyang Zhao; Yinglei Cheng; Xiaosong Shi; Xianxiang Qin; Xin Li

doi:10.3788/LOP56.052804

Laser & Optoelectronics Progress, Volume. 56, Issue 5, 052804(2019)

Terrain Classification of LiDAR Point Cloud Based on Multi-Scale Features and PointNet

Zhongyang Zhao¹, Yinglei Cheng^1、*, Xiaosong Shi¹, Xianxiang Qin¹, and Xin Li²

Author Affiliations

¹ Information and Navigation College, Air Force Engineering University, Xi'an, Shaanxi 710077, China

² School of Science, Northeast Electric Power University, Jilin, Jilin 132000, China

show less

Abstract Get PDF(in Chinese)

Figures & Tables(12)

Fig. 1. Deep neural network model combining multiscale features with PointNet

Download full size

Fig. 2. PointNet network architecture

Download full size

Fig. 3. Neighbors of different scales in point clouds. (a) Scale 1; (b) scale 2; (c) scale 3

Download full size

Fig. 4. Point cloud of Semantic 3D dataset. (a) Area 1; (b) area 2

Download full size

Fig. 5. Point cloud of Vaihingen city dataset. (a) Area 1; (b) area 2; (c) area 3

Download full size

Fig. 6. Classification results of Semantic 3D dataset. (a) Input point cloud; (b) PointNet; (c) proposed algorithm

Download full size

Fig. 7. Classification results of Vaihingen city dataset. (a) Input point cloud; (b) PointNet; (c) proposed algorithm

Download full size

Table 1. Experimental results of different scales
View table
Table 1. Experimental results of different scales
Scale s=2 s=3 s=4
Multi-scale 90.7 89.8 89.2
Scale 1 85.2 85.2 85.2
Scale 2 86.4 87.3 86.1
Scale 3 86.1 84.9 84.3

Table 2. Each category IoU of Semantic 3D dataset%

View table

Table 2. Each category IoU of Semantic 3D dataset%

Algorithm	MeanIoU	Man-madeterrain	Naturalterrain	Highvegetation	Lowvegetation	Buildings	Hardscape	Scanningartefacts	Cars
Ref. [22]	58.5	85.6	83.2	74.2	32.4	89.7	18.5	25.1	59.2
Ref. [23]	59.1	82.0	77.3	79.7	22.9	91.1	18.4	37.3	64.4
Ref. [24]	61.3	83.9	66.0	86.0	40.5	91.1	30.9	27.5	64.3
Proposed	67.4	85.6	87.1	90.5	42.3	93.2	31.6	40.8	67.8

Table 3. Classification accuracy and runtime of Semantic 3D dataset
View table
Table 3. Classification accuracy and runtime of Semantic 3D dataset
Algorithm MeanIoU /% Overallaccuracy /% Runtime /s
Ref. [22] 58.5 88.9 -
Ref. [23] 59.1 88.6 3600.00
Ref. [24] 61.3 88.1 1881.00
Proposed 67.4 90.7 4300.00

Table 4. Each category IoU of Vaihingen city dataset%
View table
Table 4. Each category IoU of Vaihingen city dataset%
Algorithm MeanIoU /% Power line Car Lowvegetation Impervioussurfaces Roof Fence /hedge Facade Shrub Tree
PointNet 32.0 0.8 23.2 32.1 47.6 84.7 2.3 5.7 15.4 76.2
Proposed 34.9 1.2 34.3 36.9 49.3 86.8 2.6 4.8 13.3 85.7

Table 5. Classification accuracy and runtime of Vaihingen city dataset
View table
Table 5. Classification accuracy and runtime of Vaihingen city dataset
Algorithm Mean IoU /% Overall accuracy /% Average class accuracy /% Runtime /s
PointNet 32.0 65.2 38.1 1500.00
Proposed 34.9 74.3 43.6 2300.00

Tools

Get Citation

Copy Citation Text

Zhongyang Zhao, Yinglei Cheng, Xiaosong Shi, Xianxiang Qin, Xin Li. Terrain Classification of LiDAR Point Cloud Based on Multi-Scale Features and PointNet[J]. Laser & Optoelectronics Progress, 2019, 56(5): 052804

Download Citation

EndNote(RIS)BibTex Plain Text

Set citation alerts for article

Save article for my favorites