Performance analysis of statistical classification algorithms in automatic extraction of single tree trunk from hand-held LiDAR point cloud

Öz

LiDAR method, whose energy is light or laser, is a measurement technique that quickly measures dense spatial data. This technique is widely used in forest areas and has an intensive data processing step. Classification comes first in the mentioned processes. Accurate detection of tree stem is an important issue in predicting tree parameters. This study was conducted to evaluate the performance of the methods used in the classification and extraction of tree stem using point clouds measured by hand-held mobile LiDAR system (HMLS). To identify the stems from the HMLS point cloud on a single-tree basis, statistical classification techniques, like logistic regression, linear discriminant analysis, random forest and support vector machine, were used. Only the points representing tree stems were classified by separating them from other parts of the trees, such as branches and leaves. It was determined that the best method was a random forest classifier based on overall accuracy results. In terms of data processing performance, a linear discriminant analysis performed better than the other methods.

Anahtar Kelimeler

• Lidar
• Point cloud
• Tree stem
• Classification

El-tipi LiDAR nokta bulutundan tek ağaç gövdesinin otomatik çıkarımında istatistiksel sınıflandırma algoritmalarının performans analizi

Öz

tekniğidir. Orman alanlarında kullanımı yaygın olan bu teknik, yoğun bir veri işleme adımına sahiptir. Bu işlemlerin en başında sınıflandırma gelir. Ağaç parametrelerinin kestiriminde ağaç gövdelerinin doğru tespiti önemli bir konudur. Bu çalışma, el-tipi mobil LiDAR (EML) ile ölçülmüş nokta bulutlarında ağaç gövde modelinin sınıflandırma ile çıkarımında kullanılan yöntemlerin performanslarını değerlendirmek amacıyla yapılmıştır. Tek ağaç bazında EML nokta bulutundan gövdenin tespit edilmesi için istatistiksel sınıflandırma tekniklerinden, lojistik regresyon, doğrusal ayrıştırma analizi, rastgele orman ve destek vektör makinesi kullanılmıştır. Sadece gövdeyi temsil eden noktalar diğer dal ve yapraklardan ayrılarak sınıflandırılmış, genel doğruluk oranına göre sınıflandırma doğruluğu en yüksek yöntem rastgele orman sınıflandırıcısı olduğu tespit edilmiştir. Veri işleme performansı açısından doğrusal ayrıştırma analizi diğer yöntemlere göre daha iyi performans sergilemiştir.

Anahtar Kelimeler

• LiDAR
• nokta bulutu
• ağaç gövdesi
• sınıflandırma

Kaynakça

1. Akar Ö, Güngör O (2015) Integrating multiple texture methods and NDVI to the Random Forest classification algorithm to detect tea and hazelnut plantation areas in northeast Turkey. International Journal of Remote Sensing, 36(2), 442-464. doi:10.1080/01431161.2014.995276
2. Bauwens S, Bartholomeus H, Calders K, Lejeune P (2016) Forest Inventory with Terrestrial LiDAR: A Comparison of Static and Hand-Held Mobile Laser Scanning. Forests, 7(12). doi:10.3390/f7060127
3. Bienert A, Georgi L, Kunz M, Maas H G, von Oheimb G (2018) Comparison and Combination of Mobile and Terrestrial Laser Scanning for Natural Forest Inventories. Forests, 9(7). doi:10.3390/f9070395
4. Bishop C M (2006) Pattern Recognition and Machine Learning: Springer.
5. Breiman L (2001) Random forests. Machine learning, 45(1), 5-32.
6. Brodu N, Lague D (2012) 3D terrestrial lidar data classification of complex natural scenes using a multi-scale dimensionality criterion: Applications in geomorphology. ISPRS Journal of Photogrammetry and Remote Sensing, 68, 121-134. doi:10.1016/j.isprsjprs.2012.01.006
7. Cabo C, Del Pozo S, Rodriguez-Gonzalvez P, Ordonez C, Gonzalez-Aguilera D (2018) Comparing Terrestrial Laser Scanning (TLS) and Wearable Laser Scanning (WLS) for Individual Tree Modeling at Plot Level. Remote Sensing, 10(4).
8. CloudCompare. (2013). Telecom ParisTech (version 2.4) [GPL software]. EDF R&D. Erişim Linki: http://www.danielgm.net/cc/

9. Dai W X, Yang B S, Liang X L, Dong Z, Huan R G, Wang Y S, Li W Y (2019) Automated fusion of forest airborne and terrestrial point clouds through canopy density analysis. ISPRS Journal of Photogrammetry and Remote Sensing, 156, 94-107. doi:10.1016/j.isprsjprs.2019.08.008
10. Dubayah R O, Drake J B (2000) Lidar Remote Sensing for Forestry. Journal of Forestry, 98, 44-52. doi:10.1093/jof/98.6.44
11. Eren E T, Düzenli T, Alpak E M (2018) The plant species used as edge elements and their usage types: The case of KTU campus. Kastamonu Üniversitesi Orman Fakültesi Dergisi, 18(2), 108-120.
12. Ghatak A (2017) Machine Learning with R: Springer Singapore.
13. Heinzel J, Ginzler C (2019) A Single-Tree Processing Framework Using Terrestrial Laser Scanning Data for Detecting Forest Regeneration. Remote Sensing, 11(1). doi:10.3390/rs11010060
14. Hyyppä E, Kukko A, Kaijaluoto R, White J C, Wulder M A, Pyörälä J, Liang X, Yu X, Wang Y, Kaartinen H, Virtanen J-P, Hyyppä J (2020) Accurate derivation of stem curve and volume using backpack mobile laser scanning. ISPRS Journal of Photogrammetry and Remote Sensing, 161, 246-262. doi:10.1016/j.isprsjprs.2020.01.018
15. James G, Witten D, Hastie T, Tibshirani R (2013) An Introduction to Statistical Learning with Applications in R.
16. James M R, Quinton J N (2014) Ultra-rapid topographic surveying for complex environments: the hand-held mobile laser scanner (HMLS). Earth Surface Processes and Landforms, 39(1), 138-142. doi:10.1002/esp.3489
17. Kirasich K, Smith T, Sadler B (2018) Random Forest vs Logistic Regression: Binary Classification for Heterogeneous Datasets. SMU Data Science Review, 1(3), 9.
18. Kuhn M, Johnson K (2013) Applied Predictive Modeling: Springer.
19. Lesmeister C (2015) Mastering Machine Learning with R: Packt.
20. Lu X, Guo Q, Li W, Flanagan J (2014) A bottom-up approach to segment individual deciduous trees using leaf-off lidar point cloud data. ISPRS Journal of Photogrammetry and Remote Sensing, 94, 1-12. doi:10.1016/j.isprsjprs.2014.03.014
21. Maltamo M, Næsset E, Vauhkonen J (2014) Forestry Applications of Airborne Laser Scanning- Concepts and Case Studies. MGM. (2019). Trabzon meteoroloji istasyonu iklim verileri. Ankara
22. Özdemir İ (2013) Yersel lazer tarama ile tek ağaç özelliklerinin belirlenmesi. Türkiye Ormancılık Dergisi, 14(1), 40-47.
23. Pal R (2017) Predictive Modeling of Drug Sensitivity: Academic Press.
24. Pessoa G G, Santos R C, Carrilho A C, Galo M, Amorim A (2019) Urban Scene Classification Using Features Extracted from Photogrammetric Point Clouds Acquired by Uav. ISPRS - International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, XLII-2/W13, 511-518. doi:10.5194/isprs-archives-XLII-2-W13-511-2019
25. Robert I. K (2015) R in Action Data analysis and graphics with R. USA: Manning Publications Co.
26. Rusu R B (2009) Semantic 3D Object Maps for Everyday Manipulation in Human Living Environments.
27. Sammartano G, Spanò A (2018) Point clouds by SLAM-based mobile mapping systems: accuracy and geometric content validation in multisensor survey and stand-alone acquisition. Applied Geomatics, 10(4), 317-339. doi:10.1007/s12518-018-0221-7
28. Tomaštík J, Saloň Š, Tunák D, Chudý F, Kardoš M (2017) Tango in forests – An initial experience of the use of the new Google technology in connection with forest inventory tasks. Computers and Electronics in Agriculture, 141, 109-117. doi:10.1016/j.compag.2017.07.015
29. Vatandaşlar C, Zeybek M (2020) Application of handheld laser scanning technology for forest inventory purposes in the NE Turkey. Turkish Journal of Agriculture and Forestry. doi:10.3906/tar-1903-40
30. Wang P, Li R H, Bu G C, Zhao R (2019) Automated low-cost terrestrial laser scanner for measuring diameters at breast height and heights of plantation trees. Plos One, 14(1). doi:10.1371/journal.pone.0209888
31. Weinmann M (2016) Reconstruction and Analysis of 3D Scenes: Springer.
32. Wilcox R R (2010) Fundamentals of Modern Statistical Methods: Springer.
33. Xiong L, Wang G Q, Bao Y, Zhou X, Wang K, Liu H L, Sun X H, Zhao R B (2019) A Rapid Terrestrial Laser Scanning Method for Coastal Erosion Studies: A Case Study at Freeport, Texas, USA. Sensors, 19(15). doi:10.3390/s19153252
34. Yener H, Koç A, Çoban H O (2006) Uzaktan Algılama Verilerinde Sınıflandırma Doğruluğunun Belirlenmesi Yöntemleri. İstanbul Üniversitesi Orman Fakültesi Dergisi, 56(2), 71-88.
35. Yrttimaa T, Saarinen N, Luoma V, Tanhuanpaa T, Kankare V, Liang X L, Hyyppa J, Holopainen M, Vastaranta M (2019) Detecting and characterizing downed dead wood using terrestrial laser scanning. ISPRS Journal of Photogrammetry and Remote Sensing, 151, 76-90. doi:10.1016/j.isprsjprs.2019.03.007
36. Zeybek M (2019) El-tipi LiDAR ölçme sistemleri ve 3B veri işleme. Türkiye Lidar Dergisi, 1(1), 10-15.
37. Zhang W M, Qi J B, Wan P, Wang H T, Xie D H, Wang X Y, Yan G J (2016) An Easy-to-Use Airborne LiDAR Data Filtering Method Based on Cloth Simulation. Remote Sensing, 8(6), 501. doi:10.3390/rs8060501