Laser & Optoelectronics Progress, Volume. 56, Issue 16, 161002(2019)
Point Cloud Classification Algorithm Based on IPTD and SVM
Figures & Tables(12)
Fig. 5. Classification results of ground points. (a) Bungalow area; (b) building area
Fig. 6. Elevation distributions of point clouds. (a) Original point clouds; (b) normalized point clouds
Fig. 8. Comparison of boundary classification results. (a) Rough classification result of bungalow area; (b) rough classification result of building area;(c) fine classification result of bungalow area; (d) fine classification result of building area
Fig. 9. Classification results of different algorithms. (a) Artificial classification results; (b) classification results of traditional SVM; (c) classification results of NN-SVM; (d) classification results of proposed algorithm
Table 1. Classification accuracy of different feature combinations%
View tableTable 1. Classification accuracy of different feature combinations%
Category Classification accuracy of unnormalized non-ground points Classification accuracy of normalized non-ground points Using all characteristics Using selected characteristics Vegetation 84.7 89.9 90.2 Building 86.5 90.5 91.6 Artificiality 48.3 77.3 77.1
Table 2. Classification accuracy of boundary region%
View tableTable 2. Classification accuracy of boundary region%
Category Results of rough classification Results of fine classification Vegetation 77.6 80.3 Building 69.4 83.1
Table 3. Classification accuracy of different algorithms
View tableTable 3. Classification accuracy of different algorithms
Algorithm Classification accuracy of every category /% Overall classification accuracy /% Time /s Traditional SVM Ground 79.2 79.6 397 Vegetation 81.2 Building 78.5 Artificiality 45.1 NN-SVM Ground 86.5 87.2 216 Vegetation 87.4 Building 88.2 Artificiality 50.5 Proposed algorithm Ground 92.7 92.6 364 Vegetation 91.6 Building 93.3 Artificiality 77.1
Tools
Get Citation
Copy Citation Text
Xiaosong Shi, Yinglei Cheng, Zhongyang Zhao, Xianxiang Qin. Point Cloud Classification Algorithm Based on IPTD and SVM[J]. Laser & Optoelectronics Progress, 2019, 56(16): 161002
Paper Information
Category: Image Processing
Received: Jan. 9, 2019
Accepted: Mar. 12, 2019
Published Online: Aug. 5, 2019
The Author Email: Xiaosong Shi (shixiaosong321@126.com)
© Copyright 2018-2021 | Chinese Laser Press.
All Rights Reserved 沪ICP备15018463号-20