Performance of Support Vector Machines and Artificial Neural Network for Mapping Endangered Tree Species Using WorldView-2 Data in Dukuduku Forest, South Africa

Endangered tree species (ETS) play a significant role in ecosystem functioning and services, land use dynamics, and other socio-economic aspects. Such aspects include ecological, economic, livelihood, and security-based and well-being benefits. The development of techniques for mapping and monitoring ETS is thus critical for understanding functioning of ecosystems. The advent of advanced imaging systems and supervised learning algorithms has provided an opportunity to map ETS over fragmenting areas. Recently, vegetation maps have been produced using advanced imaging systems such as WorldView-2 (WV-2) and robust classification algorithms such as support vector machines (SVM) and artificial neural network (ANN). However, delineation of ETS in a fragmenting ecosystem using high-resolution imagery has largely remained elusive due to the complexity of the species structure and their distribution. Therefore, the aim of the present study was to examine the utility of the advanced WV-2 data for mapping ETS in the fragmenting Dukuduku indigenous forest of South Africa using SVM and ANN classification algorithms. Specifically, the study looked at testing the advent of the additional WV-2 bands in mapping six ETS. WV-2 image was spectrally resized to separate four standard bands (SB) and four additional bands (AB). WV-2 image (8 bands: 8B) together with the SB and AB subsets was classified using SVM and ANN methods. The results showed the robustness of the two machine learning algorithms with an overall accuracy (OA) of 77.00% for SVM and 75.00% for ANN using 8B. The SB produced OA of 65.00% for SVM and 64.00% for ANN. The AB produced almost the same OA of 70.00% for both SVM and ANN. There were significant differences between the performances of the two algorithms as demonstrated by the results of McNemar's test (${\text Z}\;\text {score} \geq {1}.{96}$). This study concludes that SVM and ANN classification algorithms with WV-2 8B have the potential to map ETS in the Dukuduku indigenous forest. This study offers relatively accurate information that is important for forest managers to make informed decisions regarding management and conservation protocols of ETS.

[1]  Stefano Pignatti,et al.  Evaluating Hyperion capability for land cover mapping in a fragmented ecosystem : Pollino National Park, Italy , 2009 .

[2]  Martin Brown,et al.  Support vector machines for optimal classification and spectral unmixing , 1999 .

[3]  Belinda Reyers,et al.  Priority areas for the conservation of subtropical indigenous forest in southern Africa: a case study from KwaZulu-Natal , 2001, Biodiversity & Conservation.

[4]  Clement Atzberger,et al.  Early Detection of Bark Beetle Infestation in Norway Spruce ( Picea abies , L.) using WorldView-2 Data Frühzeitige Erkennung von Borkenkä ferbefall an Fichten mittels WorldView-2 Satellitendaten , 2014 .

[5]  Chih-Jen Lin,et al.  A Practical Guide to Support Vector Classication , 2008 .

[6]  Jun Wang,et al.  Comparison of random forest, support vector machine and back propagation neural network for electronic tongue data classification: Application to the recognition of orange beverage and Chinese vinegar , 2013 .

[7]  Mohammed N. Al-Kabi,et al.  A Comparison Study between Data Mining Tools over some Classification Methods , 2011 .

[8]  M. Cho,et al.  Assessing the effects of subtropical forest fragmentation on leaf nitrogen distribution using remote sensing data , 2013, Landscape Ecology.

[9]  Rick L. Lawrence,et al.  Mapping invasive plants using hyperspectral imagery and Breiman Cutler classifications (RandomForest) , 2006 .

[10]  A. Skidmore,et al.  Integrating imaging spectroscopy and neural networks to map grass quality in the Kruger National Park, South Africa , 2004 .

[11]  Aniruddha Ghosh,et al.  A comparison of selected classification algorithms for mapping bamboo patches in lower Gangetic plains using very high resolution WorldView 2 imagery , 2014, Int. J. Appl. Earth Obs. Geoinformation.

[12]  Paul M. Mather,et al.  The use of backpropagating artificial neural networks in land cover classification , 2003 .

[13]  J. van Staden,et al.  Screening of Zulu medicinal plants for prostaglandin-synthesis inhibitors. , 1996, Journal of ethnopharmacology.

[14]  Lars T. Waser,et al.  Evaluating the Potential of WorldView-2 Data to Classify Tree Species and Different Levels of Ash Mortality , 2014, Remote. Sens..

[15]  Ingmar Nitze,et al.  COMPARISON OF MACHINE LEARNING ALGORITHMS RANDOM FOREST, ARTIFICIAL NEURAL NETWORK AND SUPPORT VECTOR MACHINE TO MAXIMUM LIKELIHOOD FOR SUPERVISED CROP TYPE CLASSIFICATION , 2012 .

[16]  Johan A. K. Suykens,et al.  Improved Long-Term Temperature Prediction by Chaining of Neural Networks , 2001, Int. J. Neural Syst..

[17]  Clement Atzberger,et al.  Tree Species Classification with Random Forest Using Very High Spatial Resolution 8-Band WorldView-2 Satellite Data , 2012, Remote. Sens..

[18]  P. Atkinson,et al.  Introduction Neural networks in remote sensing , 1997 .

[19]  George P. Petropoulos,et al.  Support vector machines and object-based classification for obtaining land-use/cover cartography from Hyperion hyperspectral imagery , 2012, Comput. Geosci..

[20]  P. Werbos,et al.  Beyond Regression : "New Tools for Prediction and Analysis in the Behavioral Sciences , 1974 .

[21]  Xiaojun Yang,et al.  Parameterizing Support Vector Machines for Land Cover Classification , 2011 .

[22]  Ibrahim Mohamed Suliman. Eldeen,et al.  Pharmacological investigation of some trees used in South African traditional medicine. , 2005 .

[23]  J. Peñuelas,et al.  The red edge position and shape as indicators of plant chlorophyll content, biomass and hydric status. , 1994 .

[24]  John R. Jensen,et al.  Opening the black box of neural networks for remote sensing image classification , 2004 .

[25]  Helmi Zulhaidi Mohd Shafri,et al.  A comparison of support vector machine and decision tree classifications using satellite data of Langkawi Island , 2009 .

[26]  Onisimo Mutanga,et al.  Testing the reliability and stability of the internal accuracy assessment of random forest for classifying tree defoliation levels using different validation methods , 2015 .

[27]  Liu Xiang-nan Xiu Li-na,et al.  Current Status and Future Direction of the Study on Artificial Neural Network Classification Processing in Remote Sensing , 2011 .

[28]  Anthony J. Ratkowski,et al.  Validation of the QUick atmospheric correction (QUAC) algorithm for VNIR-SWIR multi- and hyperspectral imagery , 2005 .

[29]  O. Mutanga,et al.  Estimating the road edge effect on adjacent Eucalyptus grandis forests in KwaZulu-Natal, South Africa, using texture measures and an artificial neural network , 2012 .

[30]  L. S. Davis,et al.  An assessment of support vector machines for land cover classi(cid:142) cation , 2002 .

[31]  Nomzamo Bonisiwe Ndlovu,et al.  Quantifying indigenous forest change in Dukuduku from 1960 to 2008 using GIS and remote sensing techniques to support sustainable forest management planning , 2013 .

[32]  Thomas Blaschke,et al.  Land cover change assessment using decision trees, support vector machines and maximum likelihood classification algorithms , 2010, Int. J. Appl. Earth Obs. Geoinformation.

[33]  Russell G. Congalton,et al.  Assessing the accuracy of remotely sensed data : principles and practices , 1998 .

[34]  Paul M. Mather,et al.  Computer processing of remotely-sensed images , 2016 .

[35]  B. van Wyk,et al.  A broad review of commercially important southern African medicinal plants. , 2008, Journal of ethnopharmacology.

[36]  Corinna Cortes,et al.  Support-Vector Networks , 1995, Machine Learning.

[37]  K. Lee,et al.  A comparative study of artificial neural networks and support vector machines for predicting groundwater levels in a coastal aquifer , 2011 .

[38]  Benoît Stoll,et al.  Support vector machines to map rare and endangered native plants in Pacific islands forests , 2012, Ecol. Informatics.

[39]  Giles M. Foody,et al.  Multiclass and Binary SVM Classification: Implications for Training and Classification Users , 2008, IEEE Geoscience and Remote Sensing Letters.

[40]  Kristin P. Bennett,et al.  Support vector machines: hype or hallelujah? , 2000, SKDD.

[41]  George P. Petropoulos,et al.  Burnt area delineation from a uni-temporal perspective based on Landsat TM imagery classification using Support Vector Machines , 2011, Int. J. Appl. Earth Obs. Geoinformation.

[42]  Antanas Verikas,et al.  Mining data with random forests: A survey and results of new tests , 2011, Pattern Recognit..

[43]  H. Cevikalp,et al.  A new distance measure for hierarchical clustering , 2008, 2008 IEEE 16th Signal Processing, Communication and Applications Conference.

[44]  Heekuck Oh,et al.  Neural Networks for Pattern Recognition , 1993, Adv. Comput..

[45]  Leo Breiman,et al.  Random Forests , 2001, Machine Learning.

[46]  C. Ginzler,et al.  Combining ensemble modeling and remote sensing for mapping individual tree species at high spatial resolution , 2013 .

[47]  Kurt Hornik,et al.  Support Vector Machines in R , 2006 .

[48]  Mark W. Schwartz,et al.  Rare Species and Ecosystem Functioning , 2005 .

[49]  J. Benítez‐Malvido,et al.  Impact of Forest Fragmentation on Seedling Abundance in a Tropical Rain Forest , 1998 .

[50]  Michael E. Schaepman,et al.  A review on reflective remote sensing and data assimilation techniques for enhanced agroecosystem modeling , 2007, Int. J. Appl. Earth Obs. Geoinformation.

[51]  R. Pontius,et al.  Death to Kappa: birth of quantity disagreement and allocation disagreement for accuracy assessment , 2011 .

[52]  J. Privette,et al.  Inversion methods for physically‐based models , 2000 .

[53]  Lindi J. Quackenbush,et al.  INVESTIGATING NEW ADVANCES IN FOREST SPECIES CLASSIFICATION: ESTABLISHING A BASELINE , 2006 .

[54]  C. D. Rullán-Silva,et al.  Remote Monitoring of Forest Insect Defoliation -A Review- , 2013 .

[55]  Chih-Jen Lin,et al.  A comparison of methods for multiclass support vector machines , 2002, IEEE Trans. Neural Networks.

[56]  Jungho Im,et al.  Support vector machines in remote sensing: A review , 2011 .

[57]  Yih-Fang Huang,et al.  Bounds on the number of hidden neurons in multilayer perceptrons , 1991, IEEE Trans. Neural Networks.

[58]  O. Mutanga,et al.  Potential Utility of the WorldView-2 multispectral data and support vector machines algorithm to classifying land use/ cover in Dukuduku landscape, KwaZulu-Natal, South Africa , 2014 .

[59]  Charlie M. Shackleton,et al.  The importance of non-timber forest products in rural livelihood security and as safety nets: a review of evidence from South Africa , 2004 .

[60]  G. Foody.  Thematic map comparison: Evaluating the statistical significance of differences in classification accuracy , 2004 .

[61]  X. H. Zhang,et al.  Application of neural networks to identifying vegetation types from satellite images , 1997 .

[62]  Fabio Pacifici,et al.  On the relative predictive value of the new spectral bands in the WorldWiew-2 sensor , 2010, 2010 IEEE International Geoscience and Remote Sensing Symposium.

[63]  Onisimo Mutanga,et al.  Exploiting machine learning algorithms for tree species classification in a semiarid woodland using RapidEye image , 2013 .

[64]  W. Dlamini.  Multispectral detection of invasive alien plants from very high resolution 8-band satellite imagery using probabilistic graphical models , 2011 .

[65]  Lei Tian,et al.  CLASSIFICATION OF BROADLEAF AND GRASS WEEDS USING GABOR WAVELETS AND AN ARTIFICIAL NEURAL NETWORK , 2003 .

[66]  Zhengdong Zhang,et al.  Comparison of artificial neural network and support vector machine methods for urban land use/cover classifications from remote sensing images A Case Study of Guangzhou, South China , 2010, 2010 International Conference on Computer Application and System Modeling (ICCASM 2010).

[67]  Clement Atzberger,et al.  Comparative analysis of different retrieval methods for mapping grassland leaf area index using airborne imaging spectroscopy , 2015, Int. J. Appl. Earth Obs. Geoinformation.

[68]  Juergen Rossmann,et al.  Using Decision Tree Based Multiclass Support Vector Machines for Forest Mapping , 2011 .

[69]  Simon N. Stuart,et al.  2004 IUCN Red List of Threatened Species: A Global Species Assessment edited by Jonathan E.M. Baillie, Craig Hilton-Taylor & Simon N. Stuart (2004), xxiii + 191 pp., IUCN, Gland, Switzerland and Cambridge, UK. ISBN 2 8317 0826 5 (pbk), £18.50. , 2005, Oryx.

[70]  Christine W. Chan,et al.  Multiple neural networks for a long term time series forecast , 2004, Neural Computing & Applications.

[71]  Clement Atzberger,et al.  Evaluation of Sentinel-2 Spectral Sampling for Radiative Transfer Model Based LAI Estimation of Wheat, Sugar Beet, and Maize , 2011, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

[72]  Qihao Weng,et al.  A survey of image classification methods and techniques for improving classification performance , 2007 .

[73]  Giles M. Foody,et al.  Toward intelligent training of supervised image classifications: directing training data acquisition for SVM classification , 2004 .

[74]  A. Gitelson,et al.  Non-destructive determination of chlorophyll content of leaves of a green and an aurea mutant of tobacco by reflectance measurements , 1996 .

[75]  Martin Herold,et al.  Some challenges in global land cover mapping : An assessment of agreement and accuracy in existing 1 km datasets , 2008 .

[76]  Yuri Knyazikhin,et al.  Retrieval of canopy biophysical variables from bidirectional reflectance Using prior information to solve the ill-posed inverse problem , 2003 .

[77]  M. Cho,et al.  Classification of savanna tree species, in the Greater Kruger National Park region, by integrating hyperspectral and LiDAR data in a Random Forest data mining environment , 2012 .

[78]  Onisimo Mutanga,et al.  Potential utility of the spectral red-edge region of SumbandilaSat imagery for assessing indigenous forest structure and health , 2012, Int. J. Appl. Earth Obs. Geoinformation.

[79]  Giles M. Foody,et al.  Status of land cover classification accuracy assessment , 2002 .

[80]  Paul M. Mather,et al.  Support vector machines for classification in remote sensing , 2005 .

[81]  Robert A. Schowengerdt,et al.  A review and analysis of backpropagation neural networks for classification of remotely-sensed multi-spectral imagery , 1995 .

[82]  Xiaocheng Zhou,et al.  Urban Land Cover Mapping Based on Object Oriented Classification Using WorldView 2 Satellite Remote Sensing Images , 2012 .

[83]  D. Mulholland,et al.  The uterotonic activity of compounds isolated from the supercritical fluid extract of Ekebergia capensis. , 2000, Journal of pharmaceutical and biomedical analysis.

[84]  Muhammad Zafar,et al.  Medicinal Plants Inventory , 2012 .

[85]  Barnali M. Dixon,et al.  Multispectral landuse classification using neural networks and support vector machines: one or the other, or both? , 2008 .

[86]  Y. L. Everingham,et al.  Advanced satellite imagery to classify sugarcane crop characteristics , 2007, Agronomy for Sustainable Development.

[87]  D. Horler,et al.  The red edge of plant leaf reflectance , 1983 .

[88]  K. F. Wiersum,et al.  CULTIVATION OF MEDICINAL PLANTS AS A TOOL FOR BIODIVERSITY CONSERVATION AND POVERTY ALLEVIATION IN THE AMATOLA REGION, SOUTH AFRICA , 2006 .

[89]  Ana M. Cingolani,et al.  Mapping vegetation in a heterogeneous mountain rangeland using landsat data: an alternative method to define and classify land-cover units , 2004 .

[90]  John J. A. Ingram,et al.  Mapping tropical forest structure in southeastern Madagascar using remote sensing and artificial neural networks , 2005 .

[91]  A. Skidmore,et al.  Comparing accuracy assessments to infer superiority of image classification methods , 2006 .

[92]  J. Staden,et al.  Antimycobacterial activity of some trees used in South African traditional medicine , 2007 .

[93]  A. Hutchings,et al.  Zulu Medicinal Plants: An Inventory , 1996 .

[94]  P. Gessler,et al.  The multispectral separability of Costa Rican rainforest types with support vector machines and Random Forest decision trees , 2010 .

[95]  D. Roberts,et al.  Hyperspectral discrimination of tropical rain forest tree species at leaf to crown scales , 2005 .

[96]  P. J. García Nieto,et al.  Support vector machines and neural networks used to evaluate paper manufactured using Eucalyptus globulus , 2012 .

[97]  S. Ormerod,et al.  New paradigms for modelling species distributions , 2004 .

[98]  D. Everard,et al.  Classification and dynamics of a southern African subtropical coastal lowland forest , 1996 .

[99]  R. Wynne,et al.  Examining pine spectral separability using hyperspectral data from an airborne sensor: An extension of field‐based results , 2007 .

[100]  Khalid Mansour,et al.  Classifying increaser species as an indicator of different levels of rangeland degradation using WorldView-2 imagery , 2012 .

[101]  Giles M. Foody,et al.  Supervised image classification by MLP and RBF neural networks with and without an exhaustively defined set of classes , 2004 .

[102]  Jon Atli Benediktsson,et al.  Retrieval of the Height of Buildings From WorldView-2 Multi-Angular Imagery Using Attribute Filters and Geometric Invariant Moments , 2012, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

[103]  Johannes R. Sveinsson,et al.  Feature extraction for multisource data classification with artificial neural networks , 1997 .

[104]  Arnon Karnieli,et al.  redicting forest structural parameters using the image texture derived from orldView-2 multispectral imagery in a dryland forest , Israel , 2011 .