Automatic Extraction of Built-Up Areas from Very High-Resolution Satellite Imagery Using Patch-Level Spatial Features and Gestalt Laws of Perceptual Grouping

Yixiang Chen,Bo Huang,Zhiyong Lv,Yu Zhang,Pengdong Zhang

doi:10.3390/rs11243022

Abstract

Automatic extraction of built-up areas from very high-resolution (VHR) satellite images has received increasing attention in recent years. However, due to the complexity of spectral and spatial characteristics of built-up areas, it is still a challenging task to obtain their precise location and extent. In this study, a patch-based framework was proposed for unsupervised extraction of built-up areas from VHR imagery. First, a group of corner-constrained overlapping patches were defined to locate the candidate built-up areas. Second, for each patch, its salient textures and structural characteristics were represented as a feature vector using integrated high-frequency wavelet coefficients. Then, inspired by visual perception, a patch-level saliency model of built-up areas was constructed by incorporating Gestalt laws of proximity and similarity, which can effectively describe the spatial relationships between patches. Finally, built-up areas were extracted through thresholding and their boundaries were refined by morphological operations. The performance of the proposed method was evaluated on two VHR image datasets. The resulting average F-measure values were 0.8613 for the Google Earth dataset and 0.88 for the WorldView-2 dataset, respectively. Compared with existing models, the proposed method obtains better extraction results, which show more precise boundaries and preserve better shape integrity.

Highlights

Built-up areas are important land-use types, especially in urban environments
We introduce the details of our proposed method for automatic extraction of built-up areas from Very high-resolution (VHR) satellite imagery
We proposed a novel framework for unsupervised extraction of built-up areas from VHR satellite imagery

Summary

Introduction

Built-up areas are important land-use types, especially in urban environments. The precise information about the locations, extents, and distributions of built-up areas is crucial for a wide range of applications [1,2], such as land use monitoring [3], population estimation and analysis [4,5], building detection [6], urban heat island analysis [7,8], and slum and poverty monitoring [9,10]. Due to the uncertainty of object representation in feature space caused by spectral variability, the increased spatial resolution does not mean the improved accuracy for target classification [11,12], it poses a big challenge for built-up areas identification. To address this challenge, much work has been done by researchers to extract built-up areas from high-resolution satellite images. Their basic assumption that there exists high local contrast between a building and its shadow is not always true, especially in complicated scenes

Objectives

Methods

Results

Conclusion