Land Cover Classification of Multispectral LiDAR Data With an Efficient Self-Attention Capsule Network

Periodically conducting land cover mapping plays a vital role in monitoring the status and changes of the land use. The up-to-date and accurate land use database serves importantly for a wide range of applications. This letter constructs an efficient self-attention capsule network (ESA-CapsNet) for land cover classification of multispectral light detection and ranging (LiDAR) data. First, formulated with a novel capsule encoder-decoder architecture, the ESA-CapsNet performs promisingly in extracting high-level, informative, and strong feature semantics for pixel-wise land cover classification by using the five types of rasterized feature images. Furthermore, designed with a novel capsule-based attention module, the channel and spatial feature encodings are comprehensively exploited to boost the feature saliency and robustness. The ESA-CapsNet is evaluated on two multispectral LiDAR data sets and achieves an advantageous performance with the overall accuracy, average accuracy, and kappa coefficient of over 98.42%, 95.15%, and 0.9776, respectively. Comparative experiments with the existing methods also demonstrate the effectiveness and applicability of the ESA-CapsNet in land cover classification tasks.

Automated summary

In this paper, an efficient self-attention capsule network (ESA-CapsNet) is proposed for land cover classification of multispectral LiDAR data, achieving promising performance in extracting high-level feature semantics and enhancing feature saliency and robustness. The ESA-CapsNet demonstrates advantageous overall accuracy, average accuracy, and kappa coefficient, and comparative experiments validate its effectiveness and applicability in land cover classification tasks.