Integration of Multitemporal Sentinel-1 and Sentinel-2 Imagery for Land-Cover Classification Using Machine Learning Methods

Dobrinić, Dino; Medak, Damir; Gašparović, Mateo

doi:10.5194/isprs-archives-xliii-b1-2020-91-2020

Cited by 16 publications

(5 citation statements)

References 46 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Therefore, we have to separate these land cover units. For example, Du et al [63], Aslami and Ghorbani [64], Kingwell-Banham [65], Dobrinić et al [66], and Eskandari et al [67] have also been separated the irrigated lands from dry farming lands. The supervised classification was performed using SVM, ANN, MLC, MD, and Mahalanobis.…”

Section: Classification Accuracy Schemementioning

confidence: 99%

Performance Evaluation of Sentinel-2 and Landsat 8 OLI Data for Land Cover/Use Classification Using a Comparison between Machine Learning Algorithms

et al. 2021

View full text Add to dashboard Cite

With the development of remote sensing algorithms and increased access to satellite data, generating up-to-date, accurate land use/land cover (LULC) maps has become increasingly feasible for evaluating and managing changes in land cover as created by changes to ecosystem and land use. The main objective of our study is to evaluate the performance of Support Vector Machine (SVM), Artificial Neural Network (ANN), Maximum Likelihood Classification (MLC), Minimum Distance (MD), and Mahalanobis (MH) algorithms and compare them in order to generate a LULC map using data from Sentinel 2 and Landsat 8 satellites. Further, we also investigate the effect of a penalty parameter on SVM results. Our study uses different kernel functions and hidden layers for SVM and ANN algorithms, respectively. We generated the training and validation datasets from Google Earth images and GPS data prior to pre-processing satellite data. In the next phase, we classified the images using training data and algorithms. Ultimately, to evaluate outcomes, we used the validation data to generate a confusion matrix of the classified images. Our results showed that with optimal tuning parameters, the SVM classifier yielded the highest overall accuracy (OA) of 94%, performing better for both satellite data compared to other methods. In addition, for our scenes, Sentinel 2 date was slightly more accurate compared to Landsat 8. The parametric algorithms MD and MLC provided the lowest accuracy of 80.85% and 74.68% for the data from Sentinel 2 and Landsat 8. In contrast, our evaluation using the SVM tuning parameters showed that the linear kernel with the penalty parameter 150 for Sentinel 2 and the penalty parameter 200 for Landsat 8 yielded the highest accuracies. Further, ANN classification showed that increasing the hidden layers drastically reduces classification accuracy for both datasets, reducing zero for three hidden layers.

show abstract

Section: Classification Accuracy Schemementioning

confidence: 99%

Performance Evaluation of Sentinel-2 and Landsat 8 OLI Data for Land Cover/Use Classification Using a Comparison between Machine Learning Algorithms

et al. 2021

View full text Add to dashboard Cite

show abstract

“…Masiza et al [26] compared Xgboost, Random Forest (RF), Support Vector Machines (SVM), Neural Networks and Naive Bayes to classify the S1 and S2 combination in small farming areas. Dobrinić et al [27] used machine learning methods (RF and Extreme Gradient Boosting) for land cover classification in Lyon (France), increasing overall accuracy from 85% to 91%, significantly improving classification in urban areas and reducing confusions between forest and low vegetation. De Luca et al [28] studied the integration of both sensors using RF in a heterogeneous Mediterranean forest area, including time-series of each SAR and optical bands and spectral indices; they obtained an overall accuracy of 90% and the integration of SAR improved by 2.53%, which was obtained using only optical data.…”

Section: Introductionmentioning

confidence: 99%

Effect of the Synergetic Use of Sentinel-1, Sentinel-2, LiDAR and Derived Data in Land Cover Classification of a Semiarid Mediterranean Area Using Machine Learning Algorithms

2023

View full text Add to dashboard Cite

Land cover classification in semiarid areas is a difficult task that has been tackled using different strategies, such as the use of normalized indices, texture metrics, and the combination of images from different dates or different sensors. In this paper we present the results of an experiment using three sensors (Sentinel-1 SAR, Sentinel-2 MSI and LiDAR), four dates and different normalized indices and texture metrics to classify a semiarid area. Three machine learning algorithms were used: Random Forest, Support Vector Machines and Multilayer Perceptron; Maximum Likelihood was used as a baseline classifier. The synergetic use of all these sources resulted in a significant increase in accuracy, Random Forest being the model reaching the highest accuracy. However, the large amount of features (126) advises the use of feature selection to reduce this figure. After using Variance Inflation Factor and Random Forest feature importance, the amount of features was reduced to 62. The final overall accuracy obtained was 0.91 ± 0.005 (α = 0.05) and kappa index 0.898 ± 0.006 (α = 0.05). Most of the observed confusions are easily explicable and do not represent a significant difference in agronomic terms.

show abstract

“…In SNAP, the same input training datasets were given for the various classifiers like Random Forest, Minimum distance to mean, KDTree KNN, and Maximum Likelihood classifier during the classification. The classified output images of various classifiers were obtained with the help of the given input training pixels [22].…”

Section: Image Pre-processingmentioning

confidence: 99%

Accuracy Assessment of different classifiers for Sustainable Development in Landuse and Landcover mapping using Sentinel SAR and Landsat-8 data

Kanmani,

Padmanabhan,

Pari

2023

EAI Endorsed Trans Energy Web

View full text Add to dashboard Cite

Sentinel satellites make use of Synthetic Aperture Radar (SAR) which produces images with backscattered signals at fine spatial resolution from 10 m to 50 m. This study is mainly focused on evaluating and assessing the accuracy of various supervised classifiers like Random Forest classifier, Minimum Distance to mean classifier, KDTree KNN classifier, and Maximum Likelihood classifier for landuse / landcover mapping in Maduranthakam Taluk, Kancheepuram district, Tamilnadu, India. These classifiers are widely used for classifying the Sentinel SAR images. The SAR images were processed using speckle and terrain correction and converted to backscattered energy. The training datasets for the landcover classes, such as vegetation, waterbodies, settlement, and barren land, were collected from Google Earth images in high-resolution mode. These collected training datasets were given as input for the various classifiers during the classification. The obtained classified output results of various classifiers were analyzed and compared using the overall classification accuracy. The overall accuracy achieved by the Random Forest classifier for the polarization VV and VH was 92.86%, whereas the classified accuracy of various classifiers such as KDTree KNN, Minimum distance to mean, and Maximum Likelihood are found to be 81.68%, 83.17%, and 85.64% respectively. The random forest classifier yields a higher classification accuracy value due to its greater stability in allocating the pixels to the right landuse class. In order to compare and validate the results with sentinel data, the random classifier is applied with optical Landsat-8 satellite data. The classification accuracy obtained for Landsat-8 data is 84.61%. It is clearly proved that the random forest classifier with sentinel data gives the best classification accuracy results due to its high spatial resolution and spectral sensitivity. Thus accurate landuse and landcover mapping promote sustainable development by supporting decision-making at local, regional, and national levels.

show abstract

Integration of Multitemporal Sentinel-1 and Sentinel-2 Imagery for Land-Cover Classification Using Machine Learning Methods

Cited by 16 publications

References 46 publications

Performance Evaluation of Sentinel-2 and Landsat 8 OLI Data for Land Cover/Use Classification Using a Comparison between Machine Learning Algorithms

Performance Evaluation of Sentinel-2 and Landsat 8 OLI Data for Land Cover/Use Classification Using a Comparison between Machine Learning Algorithms

Effect of the Synergetic Use of Sentinel-1, Sentinel-2, LiDAR and Derived Data in Land Cover Classification of a Semiarid Mediterranean Area Using Machine Learning Algorithms

Accuracy Assessment of different classifiers for Sustainable Development in Landuse and Landcover mapping using Sentinel SAR and Landsat-8 data

Contact Info

Product

Resources

About