Road Segmentation for Remote Sensing Images using Adversarial Spatial Pyramid Networks

TL;DR

This paper introduces a novel adversarial spatial pyramid network for accurate road segmentation in remote sensing images, effectively handling complex backgrounds and limited training data through synthetic image generation and multi-scale feature extraction.

Contribution

The paper presents a new model combining structured domain adaptation with a feature pyramid network to improve road segmentation accuracy and efficiency in remote sensing images.

Findings

Achieves 78.86 IOU on Massachusetts dataset, outperforming previous methods.

Uses 4x fewer FLOPs with higher accuracy than state-of-the-art.

Effectively handles complex backgrounds and limited training data.

Abstract

Road extraction in remote sensing images is of great importance for a wide range of applications. Because of the complex background, and high density, most of the existing methods fail to accurately extract a road network that appears correct and complete. Moreover, they suffer from either insufficient training data or high costs of manual annotation. To address these problems, we introduce a new model to apply structured domain adaption for synthetic image generation and road segmentation. We incorporate a feature pyramid network into generative adversarial networks to minimize the difference between the source and target domains. A generator is learned to produce quality synthetic images, and the discriminator attempts to distinguish them. We also propose a feature pyramid network that improves the performance of the proposed model by extracting effective features from all the layers of the network for describing different scales objects. Indeed, a novel scale-wise architecture is introduced to learn from the multi-level feature maps and improve the semantics of the features. For optimization, the model is trained by a joint reconstruction loss function, which minimizes the difference between the fake images and the real ones. A wide range of experiments on three datasets prove the superior performance of the proposed approach in terms of accuracy and efficiency. In particular, our model achieves state-of-the-art 78.86 IOU on the Massachusetts dataset with 14.89M parameters and 86.78B FLOPs, with 4x fewer FLOPs but higher accuracy (+3.47% IOU) than the top performer among state-of-the-art approaches used in the evaluation.