Evaluating the effects of spatial resolution on land use and land cover classification accuracy

Mishra, Varun Narayan; Prasad, Rajendra; Kumar, Pradeep; Gupta, Dileep Kumar; Dikshit, Prabhat Kumar Singh; Dwivedi, Shridhar; Ohri, Anurag

doi:10.1109/icmoce.2015.7489727

Cited by 8 publications

(6 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The training samples' spectra were derived from Landsat data, while their land-cover labels were derived from CCI_LC. The fine classification system used in this study (Table 2) inherited that of the CCI_LC products after the removal of four mosaic land-cover types (including mosaic natural vegetation and cropland and mosaic forest and grass or shrubland) because, in the 30 m Landsat imagery, it is possible to clearly identify the mosaic land-cover types in the coarseresolution imagery (Fisher et al, 2018;Mishra et al, 2015). The three wetland land-cover types (tree-shrub-herbaceous cover, flooded, and fresh/saline or brackish water) were further combined into one wetland land-cover type as their high spatial and spectral heterogeneity as well as temporal dynamics made it difficult to identify the wetlands using remote sensing imagery (Gong et al, 2013;Ludwig et al, 2019).…”

Section: Deriving Training Samples From the Cci_lc Land-cover Productmentioning

confidence: 99%

“…Assessing the accuracy of land-cover products is an essential step in describing the quality of the products before they are used in related applications (Olofsson et al, 2013). In the past, although there has been no standard method of assessing the accuracy of land-cover maps, the error or confusion matrix has been widely considered to be the best measure (Foody and Mathur, 2004;Gómez et al, 2016;Olofsson et al, 2014).…”

Section: Accuracy Assessmentmentioning

confidence: 99%

See 1 more Smart Citation

GLC_FCS30: global land-cover product with fine classification system at 30 m using time-series Landsat imagery

Zhou

Liu

Chen

et al. 2021

Earth Syst. Sci. Data

513

195

View full text Add to dashboard Cite

Abstract. Over past decades, a lot of global land-cover products have been released; however, these still lack a global land-cover map with a fine classification system and spatial resolution simultaneously. In this study, a novel global 30 m land-cover classification with a fine classification system for the year 2015 (GLC_FCS30-2015) was produced by combining time series of Landsat imagery and high-quality training data from the GSPECLib (Global Spatial Temporal Spectra Library) on the Google Earth Engine computing platform. First, the global training data from the GSPECLib were developed by applying a series of rigorous filters to the CCI_LC (Climate Change Initiative Global Land Cover) land-cover and MCD43A4 NBAR products (MODIS Nadir Bidirectional Reflectance Distribution Function-Adjusted Reflectance). Secondly, a local adaptive random forest model was built for each 5∘×5∘ geographical tile by using the multi-temporal Landsat spectral and texture features and the corresponding training data, and the GLC_FCS30-2015 land-cover product containing 30 land-cover types was generated for each tile. Lastly, the GLC_FCS30-2015 was validated using three different validation systems (containing different land-cover details) using 44 043 validation samples. The validation results indicated that the GLC_FCS30-2015 achieved an overall accuracy of 82.5 % and a kappa coefficient of 0.784 for the level-0 validation system (9 basic land-cover types), an overall accuracy of 71.4 % and kappa coefficient of 0.686 for the UN-LCCS (United Nations Land Cover Classification System) level-1 system (16 LCCS land-cover types), and an overall accuracy of 68.7 % and kappa coefficient of 0.662 for the UN-LCCS level-2 system (24 fine land-cover types). The comparisons against other land-cover products (CCI_LC, MCD12Q1, FROM_GLC, and GlobeLand30) indicated that GLC_FCS30-2015 provides more spatial details than CCI_LC-2015 and MCD12Q1-2015 and a greater diversity of land-cover types than FROM_GLC-2015 and GlobeLand30-2010. They also showed that GLC_FCS30-2015 achieved the best overall accuracy of 82.5 % against FROM_GLC-2015 of 59.1 % and GlobeLand30-2010 of 75.9 %. Therefore, it is concluded that the GLC_FCS30-2015 product is the first global land-cover dataset that provides a fine classification system (containing 16 global LCCS land-cover types as well as 14 detailed and regional land-cover types) with high classification accuracy at 30 m. The GLC_FCS30-2015 global land-cover products produced in this paper are free access at https://doi.org/10.5281/zenodo.3986872 (Liu et al., 2020).

show abstract

Section: Deriving Training Samples From the Cci_lc Land-cover Productmentioning

confidence: 99%

Section: Accuracy Assessmentmentioning

confidence: 99%

GLC_FCS30: global land-cover product with fine classification system at 30 m using time-series Landsat imagery

Zhou

Liu

Chen

et al. 2021

Earth Syst. Sci. Data

513

195

View full text Add to dashboard Cite

show abstract

“…The choice, definition, and number of LULC classes will impact how the features on the ground are represented as well as influence the classification accuracy [29,30]. Accuracy is influenced by the sensor type [29], the spatial resolution of the image data [31], and the number of LULC classes (negatively correlated with overall accuracy [32]). A final observation is that different LULC products have differing levels of accuracy with global, continental, and national products having accuracies ranging from 66.9% to 98.0% [28].…”

Section: Related Work 21 Land Use and Land Cover Semanticsmentioning

confidence: 99%

Semantic Boosting: Enhancing Deep Learning Based LULC Classification

et al. 2021

View full text Add to dashboard Cite

The classification of land use and land cover (LULC) is a well-studied task within the domain of remote sensing and geographic information science. It traditionally relies on remotely sensed imagery and therefore models land cover classes with respect to their electromagnetic reflectances, aggregated in pixels. This paper introduces a methodology which enables the inclusion of geographical object semantics (from vector data) into the LULC classification procedure. As such, information on the types of geographic objects (e.g., Shop, Church, Peak, etc.) can improve LULC classification accuracy. In this paper, we demonstrate how semantics can be fused with imagery to classify LULC. Three experiments were performed to explore and highlight the impact and potential of semantics for this task. In each experiment CORINE LULC data was used as a ground truth and predicted using imagery from Sentinel-2 and semantics from LinkedGeoData using deep learning. Our results reveal that LULC can be classified from semantics only and that fusing semantics with imagery—Semantic Boosting—improved the classification with significantly higher LULC accuracies. The results show that some LULC classes are better predicted using only semantics, others with just imagery, and importantly much of the improvement was due to the ability to separate similar land use classes. A number of key considerations are discussed.

show abstract

“…In contrast to the previous GSPECLib that was used to store the reflectance spectra, the current GSPECLib was developed to derive training samples using the CCI_LC and MCD43A4 NBAR products. The fine classification system used in this study (Table 2) inherited that of the CCI_LC products after the removal of four mosaic land-cover types (including mosaic natural vegetation and cropland, and mosaic forest and grass or shrubland) because, in the 30-m Landsat imagery, it is possible to clearly identify the mosaic land-cover types in the coarse resolution imagery (Fisher et al, 2018;Mishra et al, 2015). The three wetland land-cover types (tree/shrub/herbaceous cover; flooded; and fresh/saline or brackish water) were further combined into one wetland land-cover type as their high spatial and spectral heterogeneity as well as temporal dynamics made it difficult to identify the wetlands using remote sensing imagery (Gong et al, 2013;Ludwig et al, 2019).…”

Section: Deriving Training Samples From the Gspeclibmentioning

confidence: 99%

GLC_FCS30: Global land-cover product with fine classification system at 30 m using time-series Landsat imagery

Zhang¹,

Liu²,

Chen³

et al. 2020

Preprint

View full text Add to dashboard Cite

Abstract. Over past decades, a lot of global land-cover products have been released, however, these is still lack of a global land-cover map with fine classification system and spatial resolution simultaneously. In this study, a novel global 30-m land-cover classification with a fine classification system for the year 2015 (GLC_FCS30-2015) was produced by combining time-series of Landsat imagery and high-quality training data from the GSPECLib (Global Spatial Temporal Spectra Library) on the Google Earth Engine computing platform. First, the global training data from the GSPECLib were developed by applying a series of rigorous filters to the MCD43A4 NBAR and CCI_LC land-cover products. Secondly, a local adaptive random forest model was built for each 5° × 5° geographical tile by using the multi-temporal Landsat spectral and textures features of the corresponding training data, and the GLC_FCS30-2015 land-cover product containing 30 land-cover types was generated for each tile. Lastly, the GLC_FCS30-2015 was validated using three different validation systems (containing different land-cover details) using 44 043 validation samples. The validation results indicated that the GLC_FCS30-2015 achieved an overall accuracy of 82.5 % and a kappa coefficient of 0.784 for the level-0 validation system (9 basic land-cover types), an overall accuracy of 71.4 % and kappa coefficient of 0.686 for the UN-LCCS (United Nations Land Cover Classification System) level-1 system (16 LCCS land-cover types), and an overall accuracy of 68.7 % and kappa coefficient of 0.662 for the UN-LCCS level-2 system (24 fine land-cover types). The comparisons against other land-cover products (CCI_LC, MCD12Q1, FROM_GLC and GlobeLand30) indicated that GLC_FCS30-2015 provides more spatial details than CCI_LC-2015 and MCD12Q1-2015 and a greater diversity of land-cover types than FROM_GLC-2015 and GlobeLand30-2010, and that GLC_FCS30-2015 achieved the best overall accuracy of 82.5% against FROM_GLC-2015 of 59.1 % and GlobeLand30-2010 of 75.9 %. Therefore, it is concluded that the GLC_FCS30-2015 product is the first global land-cover dataset that provides a fine classification system with high classification accuracy at 30 m. The GLC_FCS30-2015 global land-cover products generated in this paper is available at https://doi.org/10.5281/zenodo.3986871 (Liu et al., 2020).

show abstract

Evaluating the effects of spatial resolution on land use and land cover classification accuracy

Cited by 8 publications

References 18 publications

GLC_FCS30: global land-cover product with fine classification system at 30 m using time-series Landsat imagery

GLC_FCS30: global land-cover product with fine classification system at 30 m using time-series Landsat imagery

Semantic Boosting: Enhancing Deep Learning Based LULC Classification

GLC_FCS30: Global land-cover product with fine classification system at 30 m using time-series Landsat imagery

Contact Info

Product

Resources

About