Classifying Individual Shrub Species in UAV Images - A Case Study of the Gobi Region of Northwest China

Shrublands are the main vegetation component in the Gobi region and contribute considerably to its ecosystem. Accurately classifying individual shrub vegetation species to understand their spatial distributions and to effectively monitor species diversity in the Gobi ecosystem is essential. High-resolution remote sensing data create vegetation type inventories over large areas. However, high spectral similarity between shrublands and surrounding areas remains a challenge. In this study, we provide a case study that integrates object-based image analysis (OBIA) and the random forest (RF) model to classify shrubland species automatically. The Gobi region on the southern slope of the Tian Shan Mountains in Northwest China was analyzed using readily available unmanned aerial vehicle (UAV) RGB imagery (1.5 cm spatial resolution). Different spectral and texture index images were derived from UAV RGB images as variables for species classification. Principal component analysis (PCA) extracted features from different types of variable sets (original bands, original bands + spectral indices, and original bands + spectral indices + texture indices). We tested the ability of several non-parametric decision tree models and different types of variable sets to classify shrub species. Moreover, we analyzed three main shrubland areas comprising different shrub species and compared the prediction accuracies of the optimal model in combination with different types of variable sets. We found that the RF model could generate higher accuracy compared with the other two models. The best results were obtained using a combination of the optimal variable set and the RF model with an 88.63% overall accuracy and 0.82 kappa coefficient. Integrating OBIA and RF in the species classification process provides a promising method for automatic mapping of individual shrub species in the Gobi region and can reduce the workload of individual shrub species classification.

[1]  Alberto Gonzalez-Sanchez,et al.  Estimation of vegetation fraction using RGB and multispectral images from UAV , 2018, International Journal of Remote Sensing.

[2]  H. Wehrden,et al.  Plant communities of the southern Mongolian Gobi. , 2009 .

[3]  Xueliang Zhang,et al.  Deep learning in remote sensing applications: A meta-analysis and review , 2019, ISPRS Journal of Photogrammetry and Remote Sensing.

[4]  Asli Ozdarici-Ok,et al.  Automatic detection and delineation of citrus trees from VHR satellite imagery , 2015 .

[5]  Li Xinrong.  Study on shrub community diversity of Ordos Plateau, Inner Mongolia, Northern China , 2001 .

[6]  Juha Hyyppä,et al.  Individual Tree Detection and Classification with UAV-Based Photogrammetric Point Clouds and Hyperspectral Imaging , 2017, Remote. Sens..

[7]  Dilek Koc-San,et al.  Automatic citrus tree extraction from UAV images and digital surface models using circular Hough transform , 2018, Comput. Electron. Agric..

[8]  Max Kuhn,et al.  Building Predictive Models in R Using the caret Package , 2008 .

[9]  Martin Brandt,et al.  An unexpectedly large count of trees in the West African Sahara and Sahel , 2020, Nature.

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

[11]  R. Pu,et al.  Segmented canonical discriminant analysis of in situ hyperspectral data for identifying 13 urban tree species , 2011 .

[12]  Makoto Nakatani,et al.  An Algorithm for Estimating Chlorophyll Content in Leaves Using a Video Camera , 1998 .

[13]  Jiang Deng,et al.  Spatial pattern analysis of Haloxylon ammodendron using UAV imagery - A case study in the Gurbantunggut Desert , 2019, Int. J. Appl. Earth Obs. Geoinformation.

[14]  T. Mockler,et al.  Vegetation Index Weighted Canopy Volume Model (CVMVI) for soybean biomass estimation from Unmanned Aerial System-based RGB imagery , 2019, ISPRS Journal of Photogrammetry and Remote Sensing.

[15]  Enric Pastor,et al.  UAV Flight Experiments Applied to the Remote Sensing of Vegetated Areas , 2014, Remote. Sens..

[16]  T. Kutser,et al.  Relating Remotely Sensed Optical Variability to Marine Benthic Biodiversity , 2013, PloS one.

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

[18]  M. Onishi,et al.  Explainable identification and mapping of trees using UAV RGB image and deep learning , 2021, Scientific Reports.

[19]  Joachim Denzler,et al.  Deep learning and process understanding for data-driven Earth system science , 2019, Nature.

[20]  Jiří Doležal,et al.  Using U-Net-Like Deep Convolutional Neural Networks for Precise Tree Recognition in Very High Resolution RGB (Red, Green, Blue) Satellite Images , 2021 .

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

[22]  Lei Ma,et al.  Image Segmentation and Object-Based Image Analysis for Environmental Monitoring: Recent Areas of Interest, Researchers' Views on the Future Priorities , 2020, Remote. Sens..

[23]  Weixin Xie,et al.  A spatial-spectral SIFT for hyperspectral image matching and classification , 2019, Pattern Recognit. Lett..

[24]  Masanori Onishi,et al.  Classifying 3 Moss Species by Deep Learning, Using the “Chopped Picture” Method , 2018 .

[25]  Dongmei Chen,et al.  Segmentation for Object-Based Image Analysis (OBIA): A review of algorithms and challenges from remote sensing perspective , 2019, ISPRS Journal of Photogrammetry and Remote Sensing.

[26]  Thomas Blaschke,et al.  Object based image analysis for remote sensing , 2010 .

[27]  Fei Zhang,et al.  New research methods for vegetation information extraction based on visible light remote sensing images from an unmanned aerial vehicle (UAV) , 2019, Int. J. Appl. Earth Obs. Geoinformation.

[28]  J. Hyyppä,et al.  Range and AGC normalization in airborne discrete-return LiDAR intensity data for forest canopies , 2010 .

[29]  Masayuki Itoh,et al.  Estimating Tree Height and Diameter at Breast Height (DBH) from Digital Surface Models and Orthophotos Obtained with an Unmanned Aerial System for a Japanese Cypress (Chamaecyparis obtusa) Forest , 2017, Remote. Sens..

[30]  G. Meyer,et al.  Color indices for weed identification under various soil, residue, and lighting conditions , 1994 .

[31]  Yidan Bao,et al.  Dynamic monitoring of biomass of rice under different nitrogen treatments using a lightweight UAV with dual image-frame snapshot cameras , 2019, Plant Methods.

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

[33]  Fang Zhang,et al.  Improving land cover classification in an urbanized coastal area by random forests: The role of variable selection , 2020 .

[34]  Shrubs and species identity effects on the distribution and diversity of ground-dwelling arthropods in a Gobi desert , 2013, Journal of Insect Conservation.

[35]  A. Suratno,et al.  Tree species identification in mixed coniferous forest using airborne laser scanning , 2009 .

[36]  Huanhuan Yuan,et al.  A Comparison of Crop Parameters Estimation Using Images from UAV-Mounted Snapshot Hyperspectral Sensor and High-Definition Digital Camera , 2018, Remote. Sens..

[37]  S. Ustin,et al.  Generation of crown bulk density for Pinus sylvestris L. from lidar , 2004 .

[38]  Jose A. Jiménez-Berni,et al.  High Throughput Determination of Plant Height, Ground Cover, and Above-Ground Biomass in Wheat with LiDAR , 2018, Front. Plant Sci..

[39]  Jing Li,et al.  Estimating biomass of winter oilseed rape using vegetation indices and texture metrics derived from UAV multispectral images , 2019, Comput. Electron. Agric..

[40]  Xiaodong Yang,et al.  Modeling maize above-ground biomass based on machine learning approaches using UAV remote-sensing data , 2019, Plant Methods.

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

[42]  María Flor Álvarez-Taboada,et al.  Mapping of the Invasive Species Hakea sericea Using Unmanned Aerial Vehicle (UAV) and WorldView-2 Imagery and an Object-Oriented Approach , 2017, Remote. Sens..

[43]  S. Prasher,et al.  Application of support vector machine technology for weed and nitrogen stress detection in corn , 2006 .

[44]  N. D. Tillett,et al.  Automated Crop and Weed Monitoring in Widely Spaced Cereals , 2006, Precision Agriculture.

[45]  Gao Bing,et al.  Classification of Hyperspectral Imagery based on spectral gradient, SVM and spatial random forest , 2018, Infrared Physics & Technology.

[46]  Shulin Liu,et al.  Biomass and vegetation coverage survey in the Mu Us sandy land - based on unmanned aerial vehicle RGB images , 2021, Int. J. Appl. Earth Obs. Geoinformation.

[47]  Linhai Jing,et al.  An individual tree crown delineation method based on multi-scale segmentation of imagery , 2012 .

[48]  Sergio Marconi,et al.  Cross-site learning in deep learning RGB tree crown detection , 2020, Ecol. Informatics.

[49]  Kati J. Salovaara,et al.  Classification of Amazonian primary rain forest vegetation using Landsat ETM+ satellite imagery , 2005 .

[50]  T. Bolch,et al.  Automated detection of rock glaciers using deep learning and object-based image analysis , 2020 .

[51]  Shuhui Jiao,et al.  Improvement of Region-Merging Image Segmentation Accuracy Using Multiple Merging Criteria , 2021, Remote. Sens..

[52]  Xiaomin Tian,et al.  Classifying tree species in the plantations of southern China based on wavelet analysis and mathematical morphology , 2021, Comput. Geosci..

[53]  Gerardo Flores,et al.  Identifying and Quantifying the Abundance of Economically Important Palms in Tropical Moist Forest Using UAV Imagery , 2019, Remote. Sens..

[54]  Jungho Im,et al.  Synergistic use of QuickBird multispectral imagery and LIDAR data for object-based forest species classification , 2010 .

[55]  Baoxin Hu,et al.  Individual Urban Tree Species Classification Using Very High Spatial Resolution Airborne Multi-Spectral Imagery Using Longitudinal Profiles , 2012, Remote. Sens..

[56]  Steven E. Franklin,et al.  Pixel- and object-based multispectral classification of forest tree species from small unmanned aerial vehicles , 2018, Journal of Unmanned Vehicle Systems.

[57]  Yu Liu,et al.  Patches structure succession based on spatial point pattern features in semi-arid ecosystems of the water-wind erosion crisscross region , 2017 .

[58]  Francisco Herrera,et al.  Detection of Fir Trees (Abies sibirica) Damaged by the Bark Beetle in Unmanned Aerial Vehicle Images with Deep Learning , 2019, Remote. Sens..

[59]  Heather McNairn,et al.  Histogram-based spatio-temporal feature classification of vegetation indices time-series for crop mapping , 2018, Int. J. Appl. Earth Obs. Geoinformation.

[60]  Nikos Koutsias,et al.  Object-based classification using Quickbird imagery for delineating forest vegetation polygons in a Mediterranean test site , 2008 .

[61]  Julia Armesto,et al.  Individual Tree Detection in a Eucalyptus Plantation Using Unmanned Aerial Vehicle (UAV)-LiDAR , 2020, Remote. Sens..

[62]  Jeremy F. Wallace,et al.  Spatial statistical analysis of tree deaths using airborne digital imagery , 2013, Int. J. Appl. Earth Obs. Geoinformation.

[63]  A. Hovi,et al.  Evaluation of simulated bands in airborne optical sensors for tree species identification , 2013 .

[64]  C. Daughtry,et al.  Evaluation of Digital Photography from Model Aircraft for Remote Sensing of Crop Biomass and Nitrogen Status , 2005, Precision Agriculture.

[65]  Yuei-An Liou,et al.  Land-Use Land-Cover Classification by Machine Learning Classifiers for Satellite Observations - A Review , 2020, Remote. Sens..

[66]  S. Juutinen,et al.  Comparing ultra‐high spatial resolution remote‐sensing methods in mapping peatland vegetation , 2019, Journal of Vegetation Science.

[67]  Jan A. N. van Aardt,et al.  Extracting structural attributes from IKONOS imagery for Eucalyptus plantation forests in KwaZulu-Natal, South Africa, using image texture analysis and artificial neural networks , 2011 .

[68]  Martin Rutzinger,et al.  Dense image matching of terrestrial imagery for deriving high-resolution topographic properties of vegetation locations in alpine terrain , 2018, Int. J. Appl. Earth Obs. Geoinformation.

[69]  Simon D. Jones,et al.  The Performance of Random Forests in an Operational Setting for Large Area Sclerophyll Forest Classification , 2013, Remote. Sens..

[70]  Dehai Zhu,et al.  Large-scale crop mapping from multi-source optical satellite imageries using machine learning with discrete grids , 2021, International Journal of Applied Earth Observation and Geoinformation.

[71]  Simon Bennertz,et al.  Combining UAV-based plant height from crop surface models, visible, and near infrared vegetation indices for biomass monitoring in barley , 2015, Int. J. Appl. Earth Obs. Geoinformation.

[72]  V. Radeloff,et al.  Image texture as a remotely sensed measure of vegetation structure , 2012 .

[73]  Thomas Blaschke,et al.  Geographic Object-Based Image Analysis – Towards a new paradigm , 2014, ISPRS journal of photogrammetry and remote sensing : official publication of the International Society for Photogrammetry and Remote Sensing.

[74]  Yugyung Lee,et al.  Deep Neural Architectures for Medical Image Semantic Segmentation: Review , 2021, IEEE Access.

[75]  Jean-Michel Guldmann,et al.  Measuring continuous landscape patterns with Gray-Level Co-Occurrence Matrix (GLCM) indices: An alternative to patch metrics? , 2020 .