LPASS-Net: Lightweight Progressive Attention Semantic Segmentation Network for Automatic Segmentation of Remote Sensing Images

Abstract

Semantic segmentation of remote sensing images plays a crucial role in urban planning and development. How to perform automatic, fast, and effective semantic segmentation of considerable size and high-resolution remote sensing images has become the key to research. However, the existing segmentation methods based on deep learning are complex and often difficult to apply practically due to the high computational cost of the excessive parameters. In this paper, we propose an end-to-end lightweight progressive attention semantic segmentation network (LPASS-Net), which aims to solve the problem of reducing computational costs without losing accuracy. Firstly, its backbone features are based on a lightweight network, MobileNetv3, and a feature fusion network composed of a reverse progressive attentional feature fusion network work. Additionally, a lightweight non-local convolutional attention network (LNCA-Net) is proposed to effectively integrate global information of attention mechanisms in the spatial dimension. Secondly, an edge padding cut prediction (EPCP) method is proposed to solve the problem of splicing traces in the prediction results. Finally, evaluated on the public datasets BDCI 2017 and ISPRS Potsdam, the mIoU reaches 83.17% and 88.86%, respectively, with an inference time of 0.0271 s.

BD-YOLO: detection algorithm for high-resolution remote sensing images

“…These methods find extensive applications in various fields, including urban planning, natural resource management, agriculture, and more. Additionally, there are several traditional lightweight models, including MobileNet, ShuffleNet, and Efficientdet, which continue to be widely utilized for object detection in remote sensing images, as referenced in literature [26][27][28][29][30][31].…”