Land Cover Mapping Using Product Level Fusion Approach

Improving the accuracy of land cover maps generated using satellite imagery is a crucial task in remote sensing, since land cover maps are the primary sources in many applications. This study proposes a product level fusion approach based on different satellite-derived indices for improved land cover mapping. For this purpose, NDVI, NDWI, NDBI, and EVI indices were applied to extract the existing land cover types in Qaemshahr district using ASTER and Landsat satellite images. The extracted land cover types were then integrated to provide and improved land cover map of the study area. Accuracy of the resulting map was assessed by calculating the user, Producer, and overall accuracies and the kappa coefficient. The results show that an overall accuracy and kappa values of about 95.95% and 0.95 is achieved for the generated land cover map, which is a reliable result for this purpose. In conclusion, the study proved high performance of the proposed product level fusion approach for land cover mapping. Therefore, the output map can be used in many applications in which land cover map is an ancillary or primary data source.

کلیدواژه ها:

product level fusion ، land cover mapping ، ASTER ، Landsat