Land cover classification through Convolutional Neur-al Network model assembly: A case study of a local rural area in Thailand

Fitton, D.; Laurens, E.; Hongkarnjanakul, Natthawat; Schwob, C.; Mezeix, Laurent

doi:10.1016/j.rsase.2022.100740

Cited by 8 publications

(6 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It had higher accuracy than others, except for built-up. Although the built-up class had lower accuracy compared to other LULC classes, the prediction results of Fitton [54] showed an accuracy of 72% for the road class Radar pattern of accuracy metrics noted that ESRS performed better for cropland (Figure 5d) and grassland (Figure 5c) compared to other sampling schemes of this study; RSG was at the lowest level of precision, mainly due to their much higher sample size. For instance, SRS and RSG embraced 70.5% and 57.7% of the total reference data in grassland class, while there were no differences between the sampling schemes for built-up (Figure 5e).…”

Section: Discussionmentioning

confidence: 68%

“…For instance, SRS and RSG embraced 70.5% and 57.7% of the total reference data in grassland class, while there were no differences between the sampling schemes for built-up (Figure 5e). Inspecting the individual sampling design, accuracy metrics had a tendency toward worsening when the reference data were more randomly distributed among classes, notably for trees (Figure 5a) and built-up (Figure 5e) land uses, due to their small width relative to the tile size [54].…”

Section: Discussionmentioning

confidence: 99%

“…For example, visible rocks were mostly misclassified as barren land (see the yellow lines in Figure 6c). This land-use class is mainly considered rocks and outcrops in agricultural and natural resource terminologies of arid and semi-arid mountainous regions [54]. No vegetation can grow on rocky outcrops because it is an exposure of underlying bedrock or ancient consolidated deposits on the Earth's surface.…”

Section: Discussionmentioning

confidence: 99%

“…In our study, RSG and ESRS had reasonable accuracy based on OA and KA metrics (Table 9). Fitton [54] could not reach accuracy greater than 90% for each class using the developed convolutional neural network (CNN) model, while Jin and Mountrakis [23] provided the maximum OA of 83% for different land cover types by combining spectral, scattering and vertical structural information. Shrublands are not believed to be a major category of LULC in Iran, but their application is quite dissimilar to grassland and forest.…”

Section: Discussionmentioning

confidence: 99%

See 3 more Smart Citations

Deep Insight on Land Use/Land Cover Geospatial Assessment through Internet-Based Validation Tool in Upper Karkheh River Basin (KRB), South-West Iran

Mallah

Gorji

Balali

et al. 2023

Land

View full text Add to dashboard Cite

Recently, the demand for high-quality land use/land cover (LULC) information for near-real-time crop type mapping, in particular for multi-relief landscapes, has increased. While the LULC classes are inherently imbalanced, the statistics generally overestimate the majority classes and underestimate the minority ones. Therefore, the aim of this study was to assess the classes of the 10 m European Satellite Agency (ESA) WorldCover 2020 land use/land cover product with the support of the Google Earth Engine (GEE) in the Honam sub-basin, west Iran, using the LACOVAL (validation tool for regional-scale land cover and land cover change) online platform. The effect of imbalanced ground truth has also been explored. Four sampling schemes were employed on a total of 720 collected ground truth points over approximately 14,100 ha. The grassland and cropland totally canopied 94% of the study area, while barren land, shrubland, trees and built-up covered the rest. The results of the validation accuracy showed that the equalized sampling scheme was more realistically successful than the others in terms of roughly the same overall accuracy (91.6%), mean user’s accuracy (91.6%), mean producers’ accuracy (91.9%), mean partial portmanteau (91.9%) and kappa (0.9). The product was statistically improved to 93.5% ± 0.04 by the assembling approach and segmented with the help of supplementary datasets and visual interpretation. The findings confirmed that, in mapping LULC, data of classes should be balanced before accuracy assessment. It is concluded that the product is a reliable dataset for environmental modeling at the regional scale but needs some modifications for bare land and grassland classes in mountainous semi-arid regions of the globe.

show abstract

Section: Discussionmentioning

confidence: 68%

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

See 2 more Smart Citations

Deep Insight on Land Use/Land Cover Geospatial Assessment through Internet-Based Validation Tool in Upper Karkheh River Basin (KRB), South-West Iran

Mallah

Gorji

Balali

et al. 2023

Land

View full text Add to dashboard Cite

show abstract

“…A specialised schema for enhancing the land cover identification [ 33 ] iteratively performs the assessment of the classification errors and lowers the error rate by creating new subclasses. Neural Networks (NNs) are the driving force in achieving state-of-the-art performance for land cover mapping, with Convolutional Neural Networks (CNNs) being the most frequently used architectures [ 34 , 35 , 36 , 37 , 38 , 39 ]. Recurrent Neural Networks (RNNs), especially Long Short-Term Memory (LSTM) networks, have shown a strong potential in utilising temporal information.…”

Section: Introductionmentioning

confidence: 99%

Graph Neural Network-Based Method of Spatiotemporal Land Cover Mapping Using Satellite Imagery

Kavran

Mongus

Žalik

et al. 2023

Sensors

View full text Add to dashboard Cite

Multispectral satellite imagery offers a new perspective for spatial modelling, change detection and land cover classification. The increased demand for accurate classification of geographically diverse regions led to advances in object-based methods. A novel spatiotemporal method is presented for object-based land cover classification of satellite imagery using a Graph Neural Network. This paper introduces innovative representation of sequential satellite images as a directed graph by connecting segmented land region through time. The method’s novel modular node classification pipeline utilises the Convolutional Neural Network as a multispectral image feature extraction network, and the Graph Neural Network as a node classification model. To evaluate the performance of the proposed method, we utilised EfficientNetV2-S for feature extraction and the GraphSAGE algorithm with Long Short-Term Memory aggregation for node classification. This innovative application on Sentinel-2 L2A imagery produced complete 4-year intermonthly land cover classification maps for two regions: Graz in Austria, and the region of Portorož, Izola and Koper in Slovenia. The regions were classified with Corine Land Cover classes. In the level 2 classification of the Graz region, the method outperformed the state-of-the-art UNet model, achieving an average F1-score of 0.841 and an accuracy of 0.831, as opposed to UNet’s 0.824 and 0.818, respectively. Similarly, the method demonstrated superior performance over UNet in both regions under the level 1 classification, which contains fewer classes. Individual classes have been classified with accuracies up to 99.17%.

show abstract

Transforming urban industrial wastelands using a CNN-based land classification model

2023

Soft Comput

View full text Add to dashboard Cite

Land cover classification through Convolutional Neur-al Network model assembly: A case study of a local rural area in Thailand

Cited by 8 publications

References 35 publications

Deep Insight on Land Use/Land Cover Geospatial Assessment through Internet-Based Validation Tool in Upper Karkheh River Basin (KRB), South-West Iran

Deep Insight on Land Use/Land Cover Geospatial Assessment through Internet-Based Validation Tool in Upper Karkheh River Basin (KRB), South-West Iran

Graph Neural Network-Based Method of Spatiotemporal Land Cover Mapping Using Satellite Imagery

Transforming urban industrial wastelands using a CNN-based land classification model

Contact Info

Product

Resources

About