A hybrid deep convolutional neural network for accurate land cover classification

• Published in: International journal of applied earth observation and geoinformation Vol. 103; p. 102515
• Main Authors: Wambugu, Naftaly, Chen, Yiping, Xiao, Zhenlong, Wei, Mingqiang, Aminu Bello, Saifullahi, Marcato Junior, José, Li, Jonathan
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 01.12.2021; Elsevier
• Subjects: data collection; Deep learning; Deep supervision; Dilated convolution; hybrids; land cover; neural networks; radiometry; Remote sensing; satellites; Spatial attention; spatial data; urban development; very high spatial resolution (VHSR); Deep learning; Dilated convolution; Deep supervision; very high spatial resolution (VHSR); Spatial attention; Remote sensing
• ISSN: 1569-8432; 1872-826X
• DOI: 10.1016/j.jag.2021.102515

•Spatial attention mechanism improves delineation of various-sized objects in CNNs.•Cascading residual dilated convolution reinforces feature multi-scale inference.•Deep supervision through intermediary loss enhances the network training process.•Different components in the hybrid network can improve land cover classification. Land cover classification provides updated information regarding the Earth's resources, which is vital for agricultural investigation, urban management, and disaster monitoring. Current advances in sensor technology on satellite and aerial remote sensing (RS) devices have improved the spatial-spectral, radiometric, and temporal resolutions of images over time. These improvements offer invaluable chances of understanding land cover information. However, land cover classification from RS images is an intricate task because of the high intra-class disparities, low inter-class similarities, and image variation types. We propose a cascaded residual dilated network (CRD-Net) for land cover classification using very high spatial resolution (VHSR) images to address these challenges. The proposed hybrid network follows the encoder-decoder concept with a spatial attention block to guide the network on learnable discriminate features coupled with an intermediary loss to enhance the training process. Moreover, a cascaded residual dilated module increases the network's receptive field to enrich multi-contextual features further, thus boosting the resultant feature descriptor. Extensive experimental results demonstrate that the proposed CRD-Net outperformed state-of-the-art methods, achieving an overall accuracy (OA) of 90.73% and 90.51% on the ISPRS Potsdam land cover dataset and ISPRS Vaihingen dataset, respectively.