Mapping the Distribution of Coffee Plantations from Multi-Resolution, Multi-Temporal, and Multi-Sensor Data Using a Random Forest Algorithm

Tridawati, Anggun; Wikantika, Ketut; Susantoro, Tri Muji; Harto, Agung Budi; Darmawan, Soni; Yayusman, Lissa Fajri; Ghazali, Mochamad Firman

doi:10.3390/rs12233933

Cited by 24 publications

(14 citation statements)

References 67 publications

(122 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As reviewed in Section 2.3, TCT coefficients have not been published for atmospherically corrected S2 imagery, and Landsat's surface reflectance coefficients [96] do not cover all S2 bands. Whereas certain authors have used at-sensor-derived coefficients on surface reflectance S2 imagery [87][88][89][90][91], in a multi-temporal change application, the ever-changing effects of atmosphere result in inconsistent indices of brightness, greenness, and wetness that are directly a result of fluctuating atmospheric conditions. Although the exact effects onto index values such as dDI are not fully understood, Crist et al [42] suggests that changing atmospheric conditions will alter subsequent results derived from TCT indices.…”

Section: Discussionmentioning

confidence: 99%

“…Recently, some authors [85,86] have used coefficients developed for Landsat at-sensor reflectance products on S2 level-2A imagery. Others [87][88][89][90][91] have applied TCT coefficients to S2 level-2A data that were developed for at-sensor S2 (level-1C) imagery [43,92]. The spectral range and similarities of bands from Landsat and S2 bands are well documented [93][94][95], and the coefficients as developed by Crist [96] have been used in various studies using surface reflectance Landsat imagery [51,[97][98][99].…”

Section: Index Designmentioning

confidence: 99%

See 1 more Smart Citation

A New Application of the Disturbance Index for Fire Severity in Coastal Dunes

et al. 2021

View full text Add to dashboard Cite

Fires are a disturbance that can lead to short term dune destabilisation and have been suggested to be an initiation mechanism of a transgressive dune phase when paired with changing climatic conditions. Fire severity is one potential factor that could explain subsequent coastal dune destabilisations, but contemporary evidence of destabilisation following fire is lacking. In addition, the suitability of conventional satellite Earth Observation methods to detect the impacts of fire and the relative fire severity in coastal dune environments is in question. Widely applied satellite-derived burn indices (Normalised Burn Index and Normalised Difference Vegetation Index) have been suggested to underestimate the effects of fire in heterogenous landscapes or areas with sparse vegetation cover. This work assesses burn severity from high resolution aerial and Sentinel 2 satellite imagery following the 2019/2020 Black Summer fires on Kangaroo Island in South Australia, to assess the efficacy of commonly used satellite indices, and validate a new method for assessing fire severity in coastal dune systems. The results presented here show that the widely applied burn indices derived from NBR differentially assess vegetation loss and fire severity when compared in discrete soil groups across a landscape that experienced a very high severity fire. A new application of the Tasselled Cap Transformation (TCT) and Disturbance Index (DI) is presented. The differenced Disturbance Index (dDI) improves the estimation of burn severity, relative vegetation loss, and minimises the effects of differing soil conditions in the highly heterogenous landscape of Kangaroo Island. Results suggest that this new application of TCT is better suited to diverse environments like Mediterranean and semi-arid coastal regions than existing indices and can be used to better assess the effects of fire and potential remobilisation of coastal dune systems.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Index Designmentioning

confidence: 99%

A New Application of the Disturbance Index for Fire Severity in Coastal Dunes

et al. 2021

View full text Add to dashboard Cite

show abstract

“…The combination of information from satellite data and field data is used as a basis for preparing training data that will subsequently become input in the classification model [84]. The training data process and RF classification in this study were performed using the Vigra tool contained in the open-source software SAGA 7.6.2 [85,86], with a total amount of training data of 1135 pixels for Landsat-7 ETM+ and 1285 pixels for Landsat-8 OLI. For the initial stage, we performed tuning on several parameters, including the number of trees, the number of predictor variables (mtry), and minimum node size to obtain optimal mangrove and non-mangrove classification models.…”

Section: Random Forest Algorithmmentioning

confidence: 99%

“…For the initial stage, we performed tuning on several parameters, including the number of trees, the number of predictor variables (mtry), and minimum node size to obtain optimal mangrove and non-mangrove classification models. This study used the number of trees of as much as 100, 500, and 1000 [87][88][89], mtry = √ k (square root) and mtry = k (all variables) [45,86], and a minimum node size of 6. The application of the RF algorithm in previous studies with the parameters of the number of trees = 1000, mtry = all, and minimum node size = 6 produces an overall accuracy of 79.333% with a kappa statistic of 0.774 in mapping the distribution of coffee plantations [86].…”

Section: Random Forest Algorithmmentioning

confidence: 99%

Decision Tree and Random Forest Classification Algorithms for Mangrove Forest Mapping in Sembilang National Park, Indonesia

et al. 2022

Self Cite

View full text Add to dashboard Cite

Sembilang National Park, one of the best and largest mangrove areas in Indonesia, is very vulnerable to disturbance by community activities. Changes in the dynamic condition of mangrove forests in Sembilang National Park must be quickly and easily accompanied by mangrove monitoring efforts. One way to monitor mangrove forests is to use remote sensing technology. Recently, machine-learning classification techniques have been widely used to classify mangrove forests. This study aims to investigate the ability of decision tree (DT) and random forest (RF) machine-learning algorithms to determine the mangrove forest distribution in Sembilang National Park. The satellite data used are Landsat-7 ETM+ acquired on 30 June 2002 and Landsat-8 OLI acquired on 9 September 2019, as well as supporting data such as SPOT 6/7 image acquired in 2020–2021, MERIT DEM and an existing mangrove map. The pre-processing includes radiometric and atmospheric corrections performed using the semi-automatic classification plugin contained in Quantum GIS. We applied decision tree and random forest algorithms to classify the mangrove forest. In the DT algorithm, threshold analysis is carried out to obtain the most optimal threshold value in distinguishing mangrove and non-mangrove objects. Here, the use of DT and RF algorithms involves several important parameters, namely, the normalized difference moisture index (NDMI), normalized difference soil index (NDSI), near-infrared (NIR) band, and digital elevation model (DEM) data. The results of DT and RF classification from Landsat-7 ETM+ and Landsat-8 OLI images show similarities regarding mangrove spatial distribution. The DT classification algorithm with the parameter combination NDMI+NDSI+DEM is very effective in classifying Landsat-7 ETM+ image, while the parameter combination NDMI+NIR is very effective in classifying Landsat-8 OLI image. The RF classification algorithm with the parameter Image (6 bands), the number of trees = 100, the number of variables predictor (mtry) is square root (), and the minimum number of node sizes = 6, provides the highest overall accuracy for Landsat-7 ETM+ image, while combining Image (7 bands) + NDMI+NDSI+DEM parameters with the number of trees = 100, mtry = all variables (, and the minimum node size = 6 provides the highest overall accuracy for Landsat-8 OLI image. The overall classification accuracy is higher when using the RF algorithm (99.12%) instead of DT (92.82%) for the Landsat-7 ETM+ image, but it is slightly higher when using the DT algorithm (98.34%) instead of the RF algorithm (97.79%) for the Landsat-8 OLI image. The overall RF classification algorithm outperforms DT because all RF classification model parameters provide a higher producer accuracy in mapping mangrove forests. This development of the classification method should support the monitoring and rehabilitation programs of mangroves more quickly and easily, particularly in Indonesia.

show abstract

“…Therefore, mapping coffee plantations is critical for determining the geographic locations of coffee production hubs. When it comes to monitoring and mapping vegetation, remote sensing technology has been extensively shown to be a cost-effective, rapid, and efficient approach [10], [11]. Additionally, the use of machine learning to remote sensing analysis seeks to enhance the lives of coffee farmworkers and their families, boost the productivity of current coffee production regions, and avert either forest clearance or depletion of other natural resources [7], [12].…”

Section: Introductionmentioning

confidence: 99%

Deep Learning Model Development for Detecting Coffee Tree Changes Based on Sentinel-2 Imagery in Vietnam

Dang

Giang

et al. 2022

IEEE Access

View full text Add to dashboard Cite

Scientists and land managers have spent considerable time and resources monitoring coffee forests in the great basalt plateau. Deep learning models for coffee classification using remote sensing data have developed into a tool that may eventually replace manual image interpretation. This study proposes a U-Net model for classifying coffee planting regions using Sentinel-2 data, which aid in the annual monitoring of coffee plantation area changes. Numerous optimizer methods were evaluated and compared to support-vector-machine and random-forest methods. Twelve U-Net models were trained and compared in total. The trained deep learning models outperformed the two benchmark methods. As a result, the U-Net model with the Adadelta optimizer and 128x128x4 input data size was chosen due to its near-95 percent accuracy and 0.12 loss function value. The model was used to successfully detect location of the Vietnamese coffee ecosystem. The Net-Adadelta-128 model's output is consistent with data from statistical reports, which estimated the area of the coffee land cover to be 684, 681, and 676 thousand hectares in 2019, 2020, and 2021, respectively. The best U-Net model, which takes approximately 30 minutes to create a new classification for 55,000 square kilometres, may one day be used for coffee research and management.

show abstract

Mapping the Distribution of Coffee Plantations from Multi-Resolution, Multi-Temporal, and Multi-Sensor Data Using a Random Forest Algorithm

Cited by 24 publications

References 67 publications

A New Application of the Disturbance Index for Fire Severity in Coastal Dunes

A New Application of the Disturbance Index for Fire Severity in Coastal Dunes

Decision Tree and Random Forest Classification Algorithms for Mangrove Forest Mapping in Sembilang National Park, Indonesia

Deep Learning Model Development for Detecting Coffee Tree Changes Based on Sentinel-2 Imagery in Vietnam

Contact Info

Product

Resources

About