RSI-Net: Two-Stream Deep Neural Network for Remote Sensing Imagesbased Semantic Segmentation

TL;DR

RSI-Net is a novel two-stream deep neural network that effectively models spatial context and global information for improved semantic segmentation of remote sensing images, outperforming existing methods.

Contribution

The paper introduces RSI-Net, combining graph convolutional networks and dense atrous convolution to enhance spatial modeling and global feature extraction in RSI segmentation.

Findings

Achieved over 91% accuracy on three RSI datasets.

Outperformed six state-of-the-art methods in accuracy and F1 score.

Demonstrated effective modeling of spatial and global features.

Abstract

For semantic segmentation of remote sensing images (RSI), trade-off between representation power and location accuracy is quite important. How to get the trade-off effectively is an open question,where current approaches of utilizing very deep models result in complex models with large memory consumption. In contrast to previous work that utilizes dilated convolutions or deep models, we propose a novel two-stream deep neural network for semantic segmentation of RSI (RSI-Net) to obtain improved performance through modeling and propagating spatial contextual structure effectively and a decoding scheme with image-level and graph-level combination. The first component explicitly models correlations between adjacent land covers and conduct flexible convolution on arbitrarily irregular image regions by using graph convolutional network, while densely connected atrous convolution network (DenseAtrousCNet) with multi-scale atrous convolution can expand the receptive fields and obtain image global information. Extensive experiments are implemented on the Vaihingen, Potsdam and Gaofen RSI datasets, where the comparison results demonstrate the superior performance of RSI-Net in terms of overall accuracy (91.83%, 93.31% and 93.67% on three datasets, respectively), F1 score (90.3%, 91.49% and 89.35% on three datasets, respectively) and kappa coefficient (89.46%, 90.46% and 90.37% on three datasets, respectively) when compared with six state-of-the-art RSI semantic segmentation methods.