Estimation of Above Ground Biomass Using Support Vector Machines and ALOS/PALSAR data

被引：11

作者：

Sivasankar, Thota ^{[1
]}

Lone, Junaid Mushtaq ^{[1
]}

Sarma, K. K. ^{[1
]}

Qadir, Abdul ^{[2
]}

Raju, P. L. N. ^{[1
]}

机构：

[1] North Eastern Space Applicat Ctr, Dept Space, Umiam 793103, Meghalaya, India

[2] Univ Delaware, Dept Geog, Newark, DE 19716 USA

来源：

VIETNAM JOURNAL OF EARTH SCIENCES | 2019年 / 41卷 / 02期

关键词：

Synthetic aperture radar; ALOS-2; PALSAR-2; above ground biomass; support vector machines; ALOS PALSAR; ABOVEGROUND BIOMASS; RADAR BACKSCATTER; FOREST BIOMASS; SENSITIVITY; PLANTATION; SAR;

D O I：

10.15625/0866-7187/41/2/13690

中图分类号：

P [天文学、地球科学];

学科分类号：

07 ;

摘要：

L-band Synthetic aperture radar (SAR) data has been extensively used for forest aboveground biomass (AGB) estimation due to its higher saturation level. However, SAR backscatter is highly influenced by the topography characteristics along with the bio-geophysical properties of vegetation and underneath soil characteristics. This has limited the accuracy of directly relating the SAR backscatter with above ground biomass in highly undulated terrain. In this study, it has been observed that terrain degree of slope and aspect plays a vital role in influencing the SAR backscatter in addition with AGB. Because of this, the degree of slope and aspect along with SAR backscatter in HH (transmit and receive polarizations are horizontal) and HV (transmit horizontal and receive vertical) polarizations have been considered as inputs for Support Vector Machine (SVM) to improve the biomass retrieval accuracy. Our results demonstrate that the accuracy of AGB estimation over hilly terrain can be significantly improved by considering topographical characteristics in addition to L-band backscatter.

引用

页码：95 / 104

页数：10

共 32 条

[1] [Anonymous], 2013, Int. J. Adv. Rem. Sen., GIS and Geography
[2] Classifying Complex Mountainous Forests with L-Band SAR and Landsat Data Integration: A Comparison among Different Machine Learning Methods in the Hyrcanian Forest
Attarchi, Sara
Gloaguen, Richard
[J]. REMOTE SENSING, 2014, 6 (05) : 3624 - 3647
[3] Awad M., 2015, Efficient learning machines: Theories, concepts, and applications for engineers and system designers, P67
[4] Evaluation of ALOS/PALSAR L-Band Data for the Estimation of Eucalyptus Plantations Aboveground Biomass in Brazil
Baghdadi, Nicolas
Le Maire, Guerric
Bailly, Jean-Stephane
Ose, Kenji
Nouvellon, Yann
Zribi, Mehrez
Lemos, Cristiane
Hakamada, Rodrigo
[J]. IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING, 2015, 8 (08) : 3802 - 3811
[5] Above Ground Biomass Estimation of Dalbergia sissoo Forest Plantation from Dual-Polarized ALOS-2 PALSAR Data
Baig, Shahbaz
Qazi, Waqas A.
Akhtar, Aqeela Mobeen
Waqar, Mirza Muhammad
Ammar, Ahmad
Gilani, Hammad
Mehmood, Syed Amer
[J]. CANADIAN JOURNAL OF REMOTE SENSING, 2017, 43 (03) : 297 - 308
[6] Root biomass allocation in the world's upland forests
Cairns, MA
Brown, S
Helmer, EH
Baumgardner, GA
[J]. OECOLOGIA, 1997, 111 (01) : 1 - 11
[7] Cristianini N., 2000, An Introduction to Support Vector Machines and Other Kernel-Based Learning Methods, DOI DOI 10.1017/CBO9780511801389
[8] Evolutionary tuning of multiple SVM parameters
Friedrichs, F
Igel, C
[J]. NEUROCOMPUTING, 2005, 64 : 107 - 117
[9] Gao XX., 2012, Proceedings of the 14th International Conference on Alkali-Aggregate Reaction, P1, DOI [10.1109/RSETE.2012.6260436, DOI 10.1109/RSETE.2012.6260436]
[10] L-band ALOS PALSAR for biomass estimation of Matang Mangroves, Malaysia
Hamdan, O.
Aziz, H. Khali
Hasmadi, I. Mohd
[J]. REMOTE SENSING OF ENVIRONMENT, 2014, 155 : 69 - 78

← 1 2 3 4 →