Abstract: The development of supervised deep learning-based methods for multilabel scene classification (MLC) is one of the prominent research directions in remote sensing (RS). However, collecting ...
Abstract: Human-labeled training datasets are essential for convolutional neural networks (ConvNets) in satellite image scene classification. Annotation errors are unavoidable due to the complexity of ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results