Classification of planted forest species in southern China with airborne hyperspectral and LiDAR data

ABSTRACT Plantations in southern China are important components of forest ecosystems. However, the distribution of the native tree species is fragmented. Accurate tree species classification can provide scientific support for forest management and measurement. In this study, Gaofeng forest farm in Guangxi Province was selected as the research area, and hyperspectral image combining with LiDAR was used for tree species classification. Derivative reflectance features, vegetation indices, and spatial texture features were extracted from hyperspectral imagery and screened by random forest method. Then, these features were combined with height information provided by LiDAR used for tree species classification, using an object-based classifier. We found that the derivative reflectance feature of hyperspectral imagery was most effective for the tree species with hard leaves, the vegetation index was best suited for the tree species with dark leaves, and the texture feature was ideal for tree species with prominent veins and bright leaves. Tree height data most significantly identified taller tree species. Fusion of hyperspectral imagery with LiDAR data resulted in a maximum classification accuracy of 91.59% and a kappa coefficient of 0.897. The classification accuracy can meet the requirements of forest resource monitoring.

[1]  Robert M. Haralick,et al.  Textural Features for Image Classification , 1973, IEEE Trans. Syst. Man Cybern..

[2]  Chein-I Chang,et al.  Hyperspectral image classification and dimensionality reduction: an orthogonal subspace projection approach , 1994, IEEE Trans. Geosci. Remote. Sens..

[3]  G. Carter,et al.  Early detection of plant stress by digital imaging within narrow stress-sensitive wavebands , 1994 .

[4]  J. Roujean,et al.  Estimating PAR absorbed by vegetation from bidirectional reflectance measurements , 1995 .

[5]  F. Parmiggiani,et al.  An investigation of the textural characteristics associated with gray level cooccurrence matrix statistical parameters , 1995, IEEE Transactions on Geoscience and Remote Sensing.

[6]  Qian Du,et al.  A joint band prioritization and band-decorrelation approach to band selection for hyperspectral image classification , 1999, IEEE Trans. Geosci. Remote. Sens..

[7]  A. Gitelson,et al.  Optical Properties and Nondestructive Estimation of Anthocyanin Content in Plant Leaves¶ , 2001, Photochemistry and photobiology.

[8]  P. Gong,et al.  Analysis of in situ hyperspectral data for nutrient estimation of giant sequoia , 2002 .

[9]  William Philpot,et al.  A derivative-aided hyperspectral image analysis system for land-cover classification , 2002, IEEE Trans. Geosci. Remote. Sens..

[10]  Jing Wang,et al.  Independent component analysis-based dimensionality reduction with applications in hyperspectral image analysis , 2006, IEEE Transactions on Geoscience and Remote Sensing.

[11]  Joydeep Ghosh,et al.  An Active Learning Approach to Hyperspectral Data Classification , 2008, IEEE Transactions on Geoscience and Remote Sensing.

[12]  Patricia Gober,et al.  Per-pixel vs. object-based classification of urban land cover extraction using high spatial resolution imagery , 2011, Remote Sensing of Environment.

[13]  Dar A. Roberts,et al.  Species-Level Differences in Hyperspectral Metrics among Tropical Rainforest Trees as Determined by a Tree-Based Classifier , 2012, Remote. Sens..

[14]  L. Bruzzone,et al.  Tree species classification in the Southern Alps based on the fusion of very high geometrical resolution multispectral/hyperspectral images and LiDAR data , 2012 .

[15]  Barbara Koch,et al.  Investigating multiple data sources for tree species classification in temperate forest and use for single tree delineation , 2012, Int. J. Appl. Earth Obs. Geoinformation.

[16]  L. Alonso,et al.  A red-edge spectral index for remote sensing estimation of green LAI over agroecosystems , 2013 .

[17]  Liangpei Zhang,et al.  An SVM Ensemble Approach Combining Spectral, Structural, and Semantic Features for the Classification of High-Resolution Remotely Sensed Imagery , 2013, IEEE Transactions on Geoscience and Remote Sensing.

[18]  Hitendra Padalia,et al.  Forest tree species discrimination in western Himalaya using EO-1 Hyperion , 2014, Int. J. Appl. Earth Obs. Geoinformation.

[19]  Andreas Burkart,et al.  Generating 3D hyperspectral information with lightweight UAV snapshot cameras for vegetation monitoring: From camera calibration to quality assurance , 2015 .

[20]  Tsegaye Tadesse,et al.  Estimation of Daily Air Temperature Based on MODIS Land Surface Temperature Products over the Corn Belt in the US , 2015, Remote. Sens..

[21]  Antonio J. Plaza,et al.  Subspace-Based Support Vector Machines for Hyperspectral Image Classification , 2015, IEEE Geoscience and Remote Sensing Letters.

[22]  Lei Ma,et al.  Training set size, scale, and features in Geographic Object-Based Image Analysis of very high resolution unmanned aerial vehicle imagery , 2015 .

[23]  L. M. Moskal,et al.  Object-Based Tree Species Classification in Urban Ecosystems Using LiDAR and Hyperspectral Data , 2016 .

[24]  Hui Li,et al.  A multiple-point spatially weighted k-NN method for object-based classification , 2016, Int. J. Appl. Earth Obs. Geoinformation.

[25]  Juha Hyyppä,et al.  A comprehensive but efficient framework of proposing and validating feature parameters from airborne LiDAR data for tree species classification , 2016, Int. J. Appl. Earth Obs. Geoinformation.

[26]  Christian Wirth,et al.  The use of airborne hyperspectral data for tree species classification in a species-rich Central European forest area , 2016, Int. J. Appl. Earth Obs. Geoinformation.

[27]  Michael A. Lefsky,et al.  Review of studies on tree species classification from remotely sensed data , 2016 .

[28]  Chengming Qi,et al.  Spectral-Spatial Hyperspectral Image Classification Based on Mathematical Morphology Post-Processing , 2017, IIKI.

[29]  Qiong Cao,et al.  Optimal Decision Fusion for Urban Land-Use/Land-Cover Classification Based on Adaptive Differential Evolution Using Hyperspectral and LiDAR Data , 2017, Remote. Sens..

[30]  Neal W. Aven,et al.  Mapping urban tree species using integrated airborne hyperspectral and LiDAR remote sensing data , 2017 .

[31]  Jing Li,et al.  A Review of Wetland Remote Sensing , 2017, Sensors.

[32]  Peijun Du,et al.  A review of supervised object-based land-cover image classification , 2017 .

[33]  Deshuai Yuan,et al.  Urban Shadow Detection and Classification Using Hyperspectral Image , 2017, Journal of the Indian Society of Remote Sensing.

[34]  Xin Shen,et al.  Tree-Species Classification in Subtropical Forests Using Airborne Hyperspectral and LiDAR Data , 2017, Remote. Sens..

[35]  Vijayan K. Asari,et al.  Progressively Expanded Neural Network (PEN Net) for hyperspectral image classification: A new neural network paradigm for remote sensing image analysis , 2018, ISPRS Journal of Photogrammetry and Remote Sensing.

[36]  Qiusheng Wu,et al.  Decision-Tree, Rule-Based, and Random Forest Classification of High-Resolution Multispectral Imagery for Wetland Mapping and Inventory , 2018, Remote. Sens..

[37]  T. Matsui,et al.  Potential impact of climate change on canopy tree species composition of cool-temperate forests in Japan using a multivariate classification tree model , 2018, Ecological Research.

[38]  Seungmin Rho,et al.  Hyperspectral classification based on spectral-spatial convolutional neural networks , 2018, Eng. Appl. Artif. Intell..

[39]  Bo Du,et al.  Hyperspectral image classification via a random patches network , 2018, ISPRS Journal of Photogrammetry and Remote Sensing.

[40]  Marco Heurich,et al.  Tree species classification using plant functional traits from LiDAR and hyperspectral data , 2018, Int. J. Appl. Earth Obs. Geoinformation.

[41]  Lin Liu,et al.  Object-Based Mangrove Species Classification Using Unmanned Aerial Vehicle Hyperspectral Images and Digital Surface Models , 2018, Remote. Sens..