Building extraction from remote sensing imagery is essential for urban planning, population monitoring, and emergency response. However, traditional methods often struggle with accurate building delineation due to complex architectural features and varying imaging conditions. In this paper, we propose ABIBE, an Adaptive Building Information-Based Extraction framework that leverages Vision-Language Models (VLMs) for high-precision building extraction from remote sensing imagery. Our approach introduces two key innovations: (1) a hierarchical feature transfer mechanism that selectively extracts and adapts visual representations from the JanusPro vision-language model, and (2) a dynamic multi-scale feature fusion technique that adaptively integrates these features with spatial information through cross-modal attention mechanisms. Extensive experiments on the WHU Building Dataset demonstrate that our method significantly outperforms state-of-the-art approaches, achieving 91.39% IoU and 95.50% F1 score. We also provide comprehensive performance analysis including computational complexity, runtime efficiency, and resource requirements to evaluate the practical applicability of the proposed framework. The ABIBE framework provides a promising solution for high-precision building extraction from remote sensing imagery, with particular advantages in areas with complex architectural structures and high building density.

The large dynamic range of the satellite-borne laser altimeter receiver is a necessary condition for obtaining high-precision ranging values. The upper and lower limits of the dynamic range reflect the ecological adaptability of the altimeter. The power of the echo signal is given in the performance of the altimeter, atmospheric reflection, and target surface reflection. If the waveform of the strong echo signal with high ground reflectivity and specular reflection appears saturated, broadened and distorted, it will interfere with the extraction of helpful information during waveform inversion and affect the accuracy of ranging. Although the waveform saturation can be corrected by the mid-to-late GLAS correction method, this paper proposes a high-sensitivity signal detection technology for a space borne laser altimeter with a large dynamic range, which is suitable for laser altimeters with any linear detection system without waveform saturation. The correction can obtain a larger dynamic range, and the mechanism analysis does not require the dual-channel gain method to obtain a larger dynamic range. In this paper, the method of combining Bessel filter and adjustable gain amplifier is utilized to optimize the design bandwidth to suppress noise, and the sensitivity is not less than 5.1nW. If the parameters are not adjusted, the dynamic range is not less than 33.7dB; if the parameters are adjusted, the sensitivity is not Below 1nW, the dynamic range is about 73.9dB. Through the ground vacuum environment test, under the condition of high signal-to-noise ratio, the ranging accuracy is better than 0.08m; when the signal-to-noise ratio is not less than 8, the ranging accuracy is not less than 0.4m.

Variational optimization (VO) pansharpening methods model the imaging process of remote sensing sensors to reconstruct high-resolution multispectral (HRMS) images. However, traditional VO approaches often assume fixed degradation operators, limiting their ability to generalize across different sensors and dynamic scenarios. While model-driven deep unfolding methods partially address this issue, most rely on predefined Gaussian kernels or shallow convolution modules, which cannot accurately represent the complex and variable degradation processes of real-world images. Therefore, we propose a deep unfolding pansharpening framework with dynamic kernel prior. The framework expands the iterative solution of VO into three modules: a progressive kernel estimator, an image generator, and a prior network. First, the kernel estimator leverages Markov Chain Monte Carlo sampling and Network-based Langevin Dynamics to adaptively learn the degradation kernel. Second, the image generator unfolds the spectral and spatial fidelity subproblems to reconstruct HRMS images. Third, the prior network integrates multiscale local priors with efficient nonlocal spectral–spatial priors based on state space models to capture long-range self-similarity. Experimental results on three satellite datasets demonstrate the superiority of the proposed method.

Post-event building damage assessment from bi-temporal satellite imagery is challenging because damaged pixels are sparse, structural changes can resemble radiometric variation, and sensor changes can dominate direct feature subtraction. This study evaluates SiameseCTDM with a Sensor-Adaptive Cross-Temporal Difference Module (SA-CTDM). At each encoder stage, SA-CTDM combines absolute and magnitude-normalized relative discrepancies through a learned per-channel gate, followed by channel recalibration. The gate provides adaptive interpolation between two discrepancy definitions; it is not treated as an independently superior accuracy component. A source-preserving class-conditional objective is also evaluated for optical-to-SAR adaptation by combining supervised BRIGHT learning, continued xBD supervision, and multi-stage alignment of intact and damaged prototypes. The adaptation terms are training-only and add no inference-time parameters. On the xBD hold set, complete SA-CTDM obtains 66.68 ± 0.54% F1, compared with 66.16 ± 1.12% for original CTDM, representing a modest 0.52 percentage-point mean increase with lower variance. A normalized-only ablation is the strongest single run, so an independent gating advantage is not claimed. A unified batch-one benchmark at $512\times 512$ input resolution reports measured latency and throughput in addition to parameter counts and GFLOPs. On a fixed seed-42 sample-level BRIGHT split, the complete configuration using 611 labeled pairs reaches 34.33% F1, whereas the aligned 305-pair configuration does not outperform vanilla fine-tuning. The results therefore characterize split-specific and label-budget-dependent behavior rather than uniform few-shot superiority.

Few-shot building segmentation in remote sensing imagery remains challenging due to limited annotated data and complex geometric structures inherent in building footprints. Traditional prototype-based methods ignore rich geometric priors of buildings, leading to suboptimal feature representations and poor generalization across different geographic regions. We propose GAPL-SegNet, a novel architecture that integrates Geometry-Aware Prototype Learning with adaptive feature aggregation for superior few-shot performance. Our approach introduces several key innovations: (1) a geometric feature extractor that captures building-specific structural patterns including edges, corners, and rectangularity through dedicated detection branches; (2) a geometry-aware prototype learning framework that leverages spatial importance weighting for more discriminative prototype construction based on geometric significance; (3) a progressive adaptive training strategy with dynamic geometric loss weighting that ensures effective integration of geometric priors; and (4) systematic cross-dataset analysis quantifying domain adaptation challenges. Extensive experiments demonstrate strong performance across multiple datasets and scenarios. On the WHU Building dataset with 100 training samples, GAPL-SegNet achieves 85.75±0.43% IoU (95% CI: [85.22, 86.29]) based on five independent runs, with notable stability (CV < 0.5%), outperforming the best baseline by 3.67% IoU. Cross-dataset evaluations reveal significant challenges from resolution differences: WHU to Inria achieves 51.68% IoU (41.4% with resolution alignment) and Inria to WHU reaches 53.71% IoU (33.6% with resolution alignment). Multi-resolution analysis demonstrates that spatial resolution differences (0.3m to 1.2m) cause up to 46% IoU degradation, identifying resolution adaptation as the primary domain shift challenge that exceeds the scope of geometric priors alone. Computational efficiency analysis reveals that our method achieves a favorable performance-efficiency balance with 22.5M parameters and 5.3ms inference time. In building segmentation’s specific few-shot setting, geometric priors bring better boundary consistency and deployment efficiency compared to general-purpose foundation models.