scholarly journals Comparison of Machine Learning Methods for Mapping the Stand Characteristics of Temperate Forests Using Multi-Spectral Sentinel-2 Data

2020 ◽  
Vol 12 (18) ◽  
pp. 3019 ◽  
Author(s):  
Kourosh Ahmadi ◽  
Bahareh Kalantar ◽  
Vahideh Saeidi ◽  
Elaheh K. G. Harandi ◽  
Saeid Janizadeh ◽  
...  
Keyword(s):  
Machine Learning ◽  
Temperate Forests ◽  
Elevation Gradient ◽  
Basal Area ◽  
Forest Stand ◽  
Stem Volume ◽  
Support Vector ◽  
Stem Density ◽  
Stand Characteristics ◽  
Sentinel 2

The estimation and mapping of forest stand characteristics are vital because this information is necessary for sustainable forest management. The present study considers the use of a Bayesian additive regression trees (BART) algorithm as a non-parametric classifier using Sentinel-2A data and topographic variables to estimate the forest stand characteristics, namely the basal area (m2/ha), stem volume (m3/ha), and stem density (number/ha). These results were compared with those of three other popular machine learning (ML) algorithms, such as generalised linear model (GLM), K-nearest neighbours (KNN), and support vector machine (SVM). A feature selection was done on 28 variables including the multi-spectral bands on Sentinel-2 satellite, related vegetation indices, and ancillary data (elevation, slope, and topographic solar-radiation index derived from digital elevation model (DEM)) and then the most insignificant variables were removed from the datasets by recursive feature elimination (RFE). The study area was a mountainous forest with high biodiversity and an elevation gradient from 26 to 1636 m. An inventory dataset of 1200 sample plots was provided for training and testing the algorithms, and the predictors were fed into the ML models to compute and predict the forest stand characteristics. The accuracies and certainties of the ML models were assessed by their root mean square error (RMSE), mean absolute error (MAE), and R-squared (R2) values. The results demonstrated that BART generated the best basal area and stem volume predictions, followed by GLM, SVM, and KNN. The best RMSE values for both basal area (8.12 m2/ha) and stem volume (29.28 m3/ha) estimation were obtained by BART. Thus, the ability of the BART model for forestry application was established. On the other hand, KNN exhibited the highest RMSE values for all stand variable predictions, thereby exhibiting the least accuracy for this specific application. Moreover, the effectiveness of the narrow Sentinel-2 bands around the red edge and elevation was highlighted for predicting the forest stand characteristics. Therefore, we concluded that the combination of the Sentinel-2 products and topographic variables derived from the PALSAR data used in this study improved the estimation of the forest attributes in temperate forests.

scholarly journals Delineating Smallholder Maize Farms from Sentinel-1 Coupled with Sentinel-2 Data Using Machine Learning

2021 ◽  
Vol 13 (9) ◽  
pp. 4728
Author(s):  
Zinhle Mashaba-Munghemezulu ◽  
George Johannes Chirima ◽  
Cilence Munghemezulu
Keyword(s):  
Machine Learning ◽  
Food Security ◽  
Rural Communities ◽  
Machine Learning Algorithms ◽  
Support Vector ◽  
Subsistence Agriculture ◽  
Smallholder Farms ◽  
Main Driver ◽  
Sentinel 2

Rural communities rely on smallholder maize farms for subsistence agriculture, the main driver of local economic activity and food security. However, their planted area estimates are unknown in most developing countries. This study explores the use of Sentinel-1 and Sentinel-2 data to map smallholder maize farms. The random forest (RF), support vector (SVM) machine learning algorithms and model stacking (ST) were applied. Results show that the classification of combined Sentinel-1 and Sentinel-2 data improved the RF, SVM and ST algorithms by 24.2%, 8.7%, and 9.1%, respectively, compared to the classification of Sentinel-1 data individually. Similarities in the estimated areas (7001.35 ± 1.2 ha for RF, 7926.03 ± 0.7 ha for SVM and 7099.59 ± 0.8 ha for ST) show that machine learning can estimate smallholder maize areas with high accuracies. The study concludes that the single-date Sentinel-1 data were insufficient to map smallholder maize farms. However, single-date Sentinel-1 combined with Sentinel-2 data were sufficient in mapping smallholder farms. These results can be used to support the generation and validation of national crop statistics, thus contributing to food security.

scholarly journals Mapping of the Canopy Openings in Mixed Beech–Fir Forest at Sentinel-2 Subpixel Level Using UAV and Machine Learning Approach

2020 ◽  
Vol 12 (23) ◽  
pp. 3925
Author(s):  
Ivan Pilaš ◽  
Mateo Gašparović ◽  
Alan Novkinić ◽  
Damir Klobučar
Keyword(s):  
Machine Learning ◽  
Forest Canopy ◽  
Vegetation Index ◽  
Predictive Performance ◽  
Spatial Extent ◽  
Gradient Boosting ◽  
Support Vector ◽  
Stochastic Gradient Boosting ◽  
Extreme Gradient Boosting ◽  
Sentinel 2

The presented study demonstrates a bi-sensor approach suitable for rapid and precise up-to-date mapping of forest canopy gaps for the larger spatial extent. The approach makes use of Unmanned Aerial Vehicle (UAV) red, green and blue (RGB) images on smaller areas for highly precise forest canopy mask creation. Sentinel-2 was used as a scaling platform for transferring information from the UAV to a wider spatial extent. Various approaches to an improvement in the predictive performance were examined: (I) the highest R2 of the single satellite index was 0.57, (II) the highest R2 using multiple features obtained from the single-date, S-2 image was 0.624, and (III) the highest R2 on the multitemporal set of S-2 images was 0.697. Satellite indices such as Atmospherically Resistant Vegetation Index (ARVI), Infrared Percentage Vegetation Index (IPVI), Normalized Difference Index (NDI45), Pigment-Specific Simple Ratio Index (PSSRa), Modified Chlorophyll Absorption Ratio Index (MCARI), Color Index (CI), Redness Index (RI), and Normalized Difference Turbidity Index (NDTI) were the dominant predictors in most of the Machine Learning (ML) algorithms. The more complex ML algorithms such as the Support Vector Machines (SVM), Random Forest (RF), Stochastic Gradient Boosting (GBM), Extreme Gradient Boosting (XGBoost), and Catboost that provided the best performance on the training set exhibited weaker generalization capabilities. Therefore, a simpler and more robust Elastic Net (ENET) algorithm was chosen for the final map creation.

scholarly journals A Machine Learning Approach for Mapping Forest Vegetation in Riparian Zones in an Atlantic Biome Environment Using Sentinel-2 Imagery

2020 ◽  
Vol 12 (24) ◽  
pp. 4086
Author(s):  
Danielle Elis Garcia Furuya ◽  
João Alex Floriano Aguiar ◽  
Nayara V. Estrabis ◽  
Mayara Maezano Faita Pinheiro ◽  
Michelle Taís Garcia Furuya ◽  
...  
Keyword(s):  
Machine Learning ◽  
Environmental Planning ◽  
Riparian Zone ◽  
Learning Algorithms ◽  
Vegetation Mapping ◽  
Forest Vegetation ◽  
Machine Learning Algorithms ◽  
Riparian Zones ◽  
Support Vector ◽  
Sentinel 2

Riparian zones consist of important environmental regions, specifically to maintain the quality of water resources. Accurately mapping forest vegetation in riparian zones is an important issue, since it may provide information about numerous surface processes that occur in these areas. Recently, machine learning algorithms have gained attention as an innovative approach to extract information from remote sensing imagery, including to support the mapping task of vegetation areas. Nonetheless, studies related to machine learning application for forest vegetation mapping in the riparian zones exclusively is still limited. Therefore, this paper presents a framework for forest vegetation mapping in riparian zones based on machine learning models using orbital multispectral images. A total of 14 Sentinel-2 images registered throughout the year, covering a large riparian zone of a portion of a wide river in the Pontal do Paranapanema region, São Paulo state, Brazil, was adopted as the dataset. This area is mainly composed of the Atlantic Biome vegetation, and it is near to the last primary fragment of its biome, being an important region from the environmental planning point of view. We compared the performance of multiple machine learning algorithms like decision tree (DT), random forest (RF), support vector machine (SVM), and normal Bayes (NB). We evaluated different dates and locations with all models. Our results demonstrated that the DT learner has, overall, the highest accuracy in this task. The DT algorithm also showed high accuracy when applied on different dates and in the riparian zone of another river. We conclude that the proposed approach is appropriated to accurately map forest vegetation in riparian zones, including temporal context.

scholarly journals Combining GF-2 and Sentinel-2 Images to Detect Tree Mortality Caused by Red Turpentine Beetle during the Early Outbreak Stage in North China

Forests ◽  
2020 ◽  
Vol 11 (2) ◽  
pp. 172 ◽  
Author(s):  
Zhan ◽  
Yu ◽  
Li ◽  
Ren ◽  
Gao ◽  
...  
Keyword(s):  
Tree Mortality ◽  
Spatial Scales ◽  
Binary Classification ◽  
Insect Pest ◽  
Forest Stand ◽  
Machine Learning Algorithms ◽  
Classification And Regression Tree ◽  
Support Vector ◽  
Single Tree ◽  
Sentinel 2

In recent years, the red turpentine beetle (RTB) (Dendroctonus valens LeConte) has invaded the northern regions of China. Due to the short invasion time, the outbreak of tree mortality corresponded to a low level of damage. Important information about tree mortality, provided by remote sensing at both single-tree and forest stand scale, is needed in forest management at the early stages of outbreak. In order to detect RTB-induced tree mortality at a single-tree scale, we evaluated the classification accuracies of Gaofen-2 (GF2) imagery at different spatial resolutions (1 and 4 m) using a pixel-based method. We also simultaneously applied an object-based method to 1 m pan-sharpened images. We used Sentinel-2 (S2) imagery with different resolutions (10 and 20 m) to detect RTB-induced tree mortality and compared their classification accuracies at a larger scale—the stand scale. Three kinds of machine learning algorithms—the classification and regression tree (CART), the random forest (RF), and the support vector machine (SVM)—were applied and compared in this study. The results showed that 1 m resolution GF2 images had the highest classification accuracy using the pixel-based method and SVM algorithm (overall accuracy = 77.7%). We found that the classification of three degrees of damage percentage within the S2 pixel (0%, <15%, and 15% < x < 50%) was not successful at a forest stand scale. However, 10 m resolution S2 images could acquire effective binary classification (<15%: overall accuracy = 74.9%; 15% < x < 50%: overall accuracy = 81.0%). Our results indicated that identifying tree mortality caused by RTB at a single-tree and forest stand scale was accomplished with the combination of GF2 and S2 images. Our results are very useful for the future exploration of the patterns of spatial and temporal changes in insect pest transmission at different spatial scales.

scholarly journals Recognition of Maize Phenology in Sentinel Images with Machine Learning

Sensors ◽  
2021 ◽  
Vol 22 (1) ◽  
pp. 94
Author(s):  
Alvaro Murguia-Cozar ◽  
Antonia Macedo-Cruz ◽  
Demetrio Salvador Fernandez-Reynoso ◽  
Jorge Arturo Salgado Transito
Keyword(s):  
Machine Learning ◽  
Satellite Image ◽  
Spatial Association ◽  
Machine Learning Algorithms ◽  
Support Vector ◽  
Classification Models ◽  
Phenological Stages ◽  
Area Index ◽  
Stage Of Development ◽  
Sentinel 2

The scarcity of water for agricultural use is a serious problem that has increased due to intense droughts, poor management, and deficiencies in the distribution and application of the resource. The monitoring of crops through satellite image processing and the application of machine learning algorithms are technological strategies with which developed countries tend to implement better public policies regarding the efficient use of water. The purpose of this research was to determine the main indicators and characteristics that allow us to discriminate the phenological stages of maize crops (Zea mays L.) in Sentinel 2 satellite images through supervised classification models. The training data were obtained by monitoring cultivated plots during an agricultural cycle. Indicators and characteristics were extracted from 41 Sentinel 2 images acquired during the monitoring dates. With these images, indicators of texture, vegetation, and colour were calculated to train three supervised classifiers: linear discriminant (LD), support vector machine (SVM), and k-nearest neighbours (kNN) models. It was found that 45 of the 86 characteristics extracted contributed to maximizing the accuracy by stage of development and the overall accuracy of the trained classification models. The characteristics of the Moran’s I local indicator of spatial association (LISA) improved the accuracy of the classifiers when applied to the L*a*b* colour model and to the near-infrared (NIR) band. The local binary pattern (LBP) increased the accuracy of the classification when applied to the red, green, blue (RGB) and NIR bands. The colour ratios, leaf area index (LAI), RGB colour model, L*a*b* colour space, LISA, and LBP extracted the most important intrinsic characteristics of maize crops with regard to classifying the phenological stages of the maize cultivation. The quadratic SVM model was the best classifier of maize crop phenology, with an overall accuracy of 82.3%.

scholarly journals CARTOGRAPHY OF MOROCCAN ARGAN TREE USING COMBINED OPTICAL AND SAR IMAGERY INTEGRATED WITH DIGITAL ELEVATION MODEL

2021 ◽  
Vol XLVI-4/W5-2021 ◽  
pp. 211-217
Author(s):  
E. Elmoussaoui ◽  
A. Moumni ◽  
A. Lahrouni
Keyword(s):  
Machine Learning ◽  
Time Series ◽  
Digital Elevation Model ◽  
Optical Data ◽  
Support Vector ◽  
Ground Truth Data ◽  
Argan Tree ◽  
Digital Elevation ◽  
Elevation Model ◽  
Sentinel 2

Abstract. Forest tree species mapping became easier due to the global availability of high spatio-temporal resolution images acquired from multiple sensors. Such data can lead to better forest resources management. Machine-learning pixel based analysis was performed to multi-spectral Sentinel-2 and Synthetic Aperture Radar Sentinel-1 time series integrated with Digital Elevation Model acquired over Argan forest of Essaouira province, Morocco. The argan tree constitutes a fundamental resource for the populations of this arid area of Morocco. This research aims to use the potential of the combination of multi-sensor data to detect, map and identify argan tree from other forest species using three Machine Learning algorithms: Support Vector Machine (SVM), Maximum Likelihood (ML) and Artificial Neural Networks (ANN). The exploited datasets included Sentinel-1 (S1), Sentinel-2 (S2) time series, Shuttle Radar Topographic Missing Digital Elevation Model (DEM) layer and Ground truth data. We tested several sets of scenarios, including single S1 derived features, single S2 time series and combined S1 and S2 derived layers with DEM scene acquisition. The best results (overall accuracy OA and Kappa coefficient K) obtained from time series of optical data (NDVI): OA = 86.87%, K = 0.84, from time series of SAR data (VV+VH/VV): OA = 45.90%, K = 0.36, from the combination of optical and SAR time series (NDVI+VH+DEM): OA = 93.01%, K = 0.914, and from the fusion of optical time series and DEM layer (NDVI+DEM): OA = 93.25%, K = 0.91. These results indicate that single-sensor (S2) integrated with the DEM layer led us to obtain the highest classification results.

scholarly journals MACHINE LEARNING APPLIED TO SENTINEL-2 AND LANDSAT-8 MULTISPECTRAL AND MEDIUM-RESOLUTION SATELLITE IMAGERY FOR THE DETECTION OF RICE PRODUCTION AREAS IN NGANJUK, EAST JAVA, INDONESIA

2021 ◽  
Vol 18 (1) ◽  
pp. 19
Author(s):  
Terry Devara ◽  
Arie Wahyu Wijayanto
Keyword(s):  
Machine Learning ◽  
Composite Index ◽  
Remote Sensing Data ◽  
Rice Production ◽  
Support Vector ◽  
Landsat 8 ◽  
Sampling Frame ◽  
Detection Model ◽  
Production Areas ◽  
Sentinel 2

Statistics Indonesia (BPS) has been introducing the use of Area Sampling Frame (ASF) surveys from 2018 to estimate rice production areas, although the process continues to suffer from the high costs of human and other resources. To support this type of conventional field survey, a more scalable and inexpensive approach using publicly-available remote sensing data, for example from the Sentinel-2 and Landsat-8 satellites, has been explored. In this research, we compare the performance gain from Sentinel-2 and Landsat-8 images using a multiple composite-index enriched machine learning classifier to detect rice production areas located in Nganjuk, East Java, Indonesia as a case study area. We build a detection model from a set of machine learning classifiers, Decision Tree (CART), Support Vector Machine, Logistic Regression, Ensemble Bagging Methods (Random Forest and Extra Trees), and Ensemble Boosting Methods (AdaBoost and XGBoost). The composite indices consist of the NDVI and EVI for agricultural and forest areas, NDWI for water and cloud, and NDBI, NDTI, and BSI for built-up areas, fallows, and asphalt-based roads. Validated by k-fold cross-validation, Sentinel-2 and Landsat-8 achieved F1-scores of 0.930 and 0.919 respectively at the scale of 30 meters per pixel. Using a 10 meter resolution per pixel for the Sentinel-2 imagery showed an increased F1-score of up to 0.971. Our evaluation shows that the higher spatial resolution imagery of Sentinel-2 achieves a better prediction, not only performance-wise, but also as a better representation of actual conditions.

Secondary forest succession in a tropical dry forest: patterns of development across a 50-year chronosequence in lowland Bolivia

2002 ◽  
Vol 18 (1) ◽  
pp. 53-66 ◽  
Author(s):  
DEBORAH K. KENNARD
Keyword(s):  
Species Richness ◽  
Tree Species ◽  
Stand Structure ◽  
Secondary Forest ◽  
Tropical Dry Forest ◽  
Basal Area ◽  
Forest Stand ◽  
Dry Forest ◽  
Stem Density ◽  
Mature Forest

Stand structure, species richness and population structures of tree species were characterized in 12 stands representing 50 y of succession following slash-and-burn agriculture in a tropical dry forest in lowland Bolivia. Estimates of tree species richness, canopy cover and basal area reached or surpassed 75% of mature forest levels in the 5-, 8-, and 23-y-old stands respectively. Total stem density of the 50-y-old stand was almost twice that of the mature forest stand. This rapid recovery may be due to a high percentage of sprouting tree species, potentially high seed fall into abandoned fields, or the disturbance history of the mature stand. The even-aged size-class structures, dominance of long-lived pioneers, and presence of charcoal and pottery shards in soils of the mature forest stand suggest it formed after a severe disturbance, possibly fire of anthropogenic origin.

scholarly journals Recherches sur la structure des peuplements vierges de hêtre oriental (Fagus orientalis Lipsky), dans les futaies du Nord de l'Iran

1973 ◽  
Vol 38 ◽  
Author(s):  
A. Asli ◽  
S. T. Nedialkov
Keyword(s):  
Basal Area ◽  
Yield Management ◽  
Stem Volume ◽  
Basic Information ◽  
Optimum Conditions ◽  
Northern Iran ◽  
Fagus Orientalis ◽  
Fagus Orientalis Lipsky ◽  
Structure Des Peuplements ◽  
Stand Characteristics

Research in stand of virgin community of beech 'Fagus orientalis Lipsky in the  Caspian forests of the northern Iran - The forest  communities of Fagetum ilexesumof the northern slopes of the Elborz Mountains are studied and  analysed. Density, basal area, height and stem volume are studied as  ecological characteristics. Data are presented by curves, tables and figures.      Comparisons are made between present stand characteristics and those of  optimum conditions for growth of an uneven-aged forest.     Data presented provide the basic information necessary to optimize  sustained yield management of the Caspian forests.

scholarly journals WIDE-AREA MAPPING OF FOREST WITH NATIONAL AIRBORNE LASER SCANNING AND FIELD INVENTORY DATASETS

2016 ◽  
Vol XLI-B8 ◽  
pp. 727-731 ◽  
Author(s):  
J.-M. Monnet ◽  
C. Ginzler ◽  
J.-C. Clivaz
Keyword(s):  
Laser Scanning ◽  
Prediction Models ◽  
Basal Area ◽  
National Forest Inventory ◽  
Airborne Laser Scanning ◽  
Stem Volume ◽  
Stem Density ◽  
Maximum Height ◽  
Airborne Laser

Airborne laser scanning (ALS) remote sensing data are now available for entire countries such as Switzerland. Methods for the estimation of forest parameters from ALS have been intensively investigated in the past years. However, the implementation of a forest mapping workflow based on available data at a regional level still remains challenging. A case study was implemented in the Canton of Valais (Switzerland). The national ALS dataset and field data of the Swiss National Forest Inventory were used to calibrate estimation models for mean and maximum height, basal area, stem density, mean diameter and stem volume. When stratification was performed based on ALS acquisition settings and geographical criteria, satisfactory prediction models were obtained for volume (R&lt;sup&gt;2&lt;/sup&gt;&thinsp;=&thinsp;0.61 with a root mean square error of 47&thinsp;%) and basal area (respectively 0.51 and 45&thinsp;%) while height variables had an error lower than 19%. This case study shows that the use of nationwide ALS and field datasets for forest resources mapping is cost efficient, but additional investigations are required to handle the limitations of the input data and optimize the accuracy.

Sign in / Sign up

Export Citation Format

Share Document