Remote Sensing Scene Classification using ConvNeXt-Tiny Model with Attention Mechanismand Label Smoothing

Remote Sensing Scene Classification using ConvNeXt-Tiny Model with Attention Mechanismand Label Smoothing

remote sensing scene classification;deep learning; ConvNeXt-Tiny, ECANet; label smoothing regularization

Remote Sensing Scene Classification (RSSC) is the discrete categorization of remote sensing images into various classes of scene categories based on their image content. RSSC plays an important role in many fields, such as agriculture, land mapping, and identification of disaster-prone areas. Therefore, a reliable and accurate RSSC algorithm is required to ensure the accuracy of land identification. Many existing studies in recent years have used deep learning methods, especially CNN combined with attentionmodules to solve this problem. This study focuses on solving the RSSC problem by proposing a deep learning-based method (CNN) with a ConvNeXt-Tiny model integrated with the Efficient Channel AttentionModule (ECANet)and label smoothing regularization (LSR).The ConvNeXt-Tiny model shows that a persistent superior outperforms the ‘large’ model in convinced metrics.ConvNeXt-Tiny model also has a huge advantage in high-precision positioning and higher classification accuracy and localization precision in a variety of complicated scenarios of remote sensing scene recognition.Experiments in this study also aim to prove that the integration of the attention module and LSR in the basic CNN network can improve accuracy because the attention module can strengthen important features and weaken features that are less useful for classification. The experimental results proved that the integration of ECANetand LSR in the ConvNeXt-Tiny base network obtained a higher accuracyof 0.38% in the UC-Merced dataset, 0.7% in the AID, and 0.4% in the WHU-RS19dataset than the ConvNeXt-Tiny model without ECANetand LSR. The ConvNeXt-Tiny model with ECANetintegration and LSR obtained an Accuracyof 99.00±0.41% in the UC-Merced dataset, 95.08±0.20% in AID, and 99.50±0.31% in the WHU-RS19 dataset

Rachmawan Atmaji Perdana1*, Aniati Murni Arimurthy2, Risnandar

https://jurnal.iaii.or.id/index.php/RESTI/article/view/5731/940

Computer Science, Faculty of Computer Science, University of Indonesia, Depok, Indonesia

21-06-2024

FAJAR BAGUS W

PDF

ENGLISH

TEXT