Advances and Future Prospects in Building Extraction From High-Resolution Remote Sensing Images

Abstract

Automatic building extraction from high resolution remains challenging for wide applications. Previous studies have reviewed building extraction methods, but rapid advances in deep learning (DL) highlight the need for an updated review focused on high-resolution remote sensing image (HRSI) applications. Therefore, we reviewed 243 papers to identify future research trends and highlight five key challenges in HRSI-based building extraction: complex boundary optimization, shape and spectrum variability, insufficient samples, tree and shadow interference, and lightweight model applications. Subsequently, we explore five corresponding opportunities leveraging advanced DL techniques. Existing methods are then categorized into two main groups: nondeep-learning-based and DL-based approaches. These are further subdivided into four broad categories: traditional methods, machine learning-based methods, semantic segmentation-based methods, and vector-based methods, which are discussed in detail based on their underlying principles. In addition, we present several practical applications and introduce ten publicly available benchmarks. We conducted a comprehensive evaluation of nine representative methods alongside our proposed Multiple-parallel vision Mamba network (MVMNet) using three public datasets: WHU, Massachusetts, and WHU Satellite I building datasets. Among these methods, HD-Net and MSSDMAP-Net demonstrated superior performance in addressing challenges associated with complex boundaries, while BuildFormerSegDP and CBRNet showed enhanced effectiveness in mitigating the impact of tree occlusion. Notably, MVMNet outperforms the other methods in addressing the challenges associated with spectral heterogeneity within buildings, achieving optimal intersection over union values of 0.9076, 0.7376, and 0.6612, respectively.

Keywords

• Building extraction
• deep learning (DL)
• high-resolution
• overview