Remote Sensing Image Road Extraction Network Based on MSPFE-Net

Road extraction is a hot task in the field of remote sensing, and it has been widely concerned applied by researchers, especially using deep learning methods. However, many models convolutional neural networks ignore attributes roads, shape road banded discrete. In addition, continuity accuracy are also affected narrow roads blocked trees. This paper designs network (MSPFE-Net) based on multi-level strip pooling feature enhancement. The overall architecture MSPFE-Net encoder-decoder, this two main modules. One module, which aggregates long-range dependencies different levels to ensure connectivity road. other module enhancement used enhance clarity local details We perform series experiments dataset, Massachusetts Roads Dataset, public dataset. experimental data showed that model was better than comparison models.