
The EcoCropsAID dataset, comprising aerial images of five key economic crops in Thailand, presents significant challenges for land use classification due to variations in image quality and similarities between different crop categories, creating opportunities for developing novel deep learning algorithms.


coremsg

ecocropsaid-a-challenging-dataset-for-economic-crop-classification-from-aerial-images-with-diverse-patterns-and-similarities-across-categories


EcoCropsAID: A Challenging Dataset for Economic Crop Classification from Aerial Images with Diverse Patterns and Similarities Across Categories


title_rewrite


This paper introduces a novel deep learning model, SpecSAR-Former, and a new dataset, Dynamic World+, for more accurate and efficient global land use and land cover (LULC) mapping by integrating Sentinel-1 and Sentinel-2 satellite data.


specsar-former-a-lightweight-transformer-based-network-for-global-land-use-and-land-cover-mapping-using-integrated-sentinel-1-and-sentinel-2-data


SpecSAR-Former: A Lightweight Transformer-Based Network for Global Land Use and Land Cover Mapping Using Integrated Sentinel-1 and Sentinel-2 Data
