Improvement in the Accuracy of the Postclassification of Land Use and Land Cover Using Landsat 8 Data Based on the Majority of Segment-Based Filtering Approach

<div>Determination of training samples and reference map of the study area: (a, c) annual mosaic of SPOT 6/7 images in 2019; (b) compilation of the Grid Feature Index (GIF) with dimensions of 2 km × 2 km is carried out to determine training samples and LULC. references; (d) centre point of each GIF block buffered with a distance of 200 m is used as the location for training samples and references from LULC classes.</div>

The Scientific World Journal

fig3

Figure 3

Figure 3: Improvement in the Accuracy of the Postclassification of Land Use and Land Cover Using Landsat 8 Data Based on the Majority of Segment-Based Filtering Approach 

Figure 3 | Improvement in the Accuracy of the Postclassification of Land Use and Land Cover Using Landsat 8 Data Based on the Majority of Segment-Based Filtering Approach 