Fully Convolutional Networks for Dense Semantic Labelling of High-Resolution Aerial Imagery

TL;DR

This paper demonstrates that fully convolutional neural networks can effectively perform dense semantic labeling of high-resolution aerial imagery, achieving state-of-the-art accuracy without the need for deconvolution or interpolation.

Contribution

It adapts fully convolutional networks to overhead imagery and introduces a hybrid fine-tuning approach for improved semantic labeling accuracy.

Findings

Achieved state-of-the-art results on ISPRS Vaihingen and Potsdam datasets.

No downsampling or deconvolution needed for full-resolution labeling.

Fine-tuning pre-trained CNNs enhances performance.

Abstract

The trend towards higher resolution remote sensing imagery facilitates a transition from land-use classification to object-level scene understanding. Rather than relying purely on spectral content, appearance-based image features come into play. In this work, deep convolutional neural networks (CNNs) are applied to semantic labelling of high-resolution remote sensing data. Recent advances in fully convolutional networks (FCNs) are adapted to overhead data and shown to be as effective as in other domains. A full-resolution labelling is inferred using a deep FCN with no downsampling, obviating the need for deconvolution or interpolation. To make better use of image features, a pre-trained CNN is fine-tuned on remote sensing data in a hybrid network context, resulting in superior results compared to a network trained from scratch. The proposed approach is applied to the problem of labelling high-resolution aerial imagery, where fine boundary detail is important. The dense labelling yields state-of-the-art accuracy for the ISPRS Vaihingen and Potsdam benchmark data sets.