Hongbo Zhu

Accurate exploitation of spatial structures and spectral characteristics is essential for fine-grained crop classification using remote sensing imagery. Although multi-source remote sensing data provide complementary information, most existing methods implicitly assume homogeneous data sources with consistent spatial resolution. In practice, high spatial resolution and rich spectral information are usually provided by different sensors, making cross-source spatial–spectral fusion a non-trivial challenge. To address this issue, we propose SSF-TransUnet, a dual-branch spatial–spectral joint modeling framework for fine crop classification. The proposed network explicitly decouples spatial structure extraction and spectral discriminability learning by jointly utilizing high spatial resolution imagery and multi-spectral observations acquired from different satellite sensors within a unified architecture. To support model training and evaluation, we construct SSCR-Agri, a spatial–spectral complementary resolution agricultural dataset integrating meter-level GF-2 imagery and multi-spectral Sentinel-2 data from five representative agricultural regions in northern China, covering five crop categories including corn, rice, wheat, potato, and others. Extensive experiments demonstrate that SSF-TransUnet consistently outperforms representative CNN-based and hybrid CNN–Transformer models. The proposed method achieves an overall accuracy (OA) of 81.84% and a mean Intersection over Union (mIoU) of 0.6954 in fine-grained crop classification, effectively distinguishing crops. These results highlight the effectiveness of spatial–spectral joint modeling for high-resolution crop mapping and demonstrate its potential for precision agriculture and large-scale agricultural monitoring applications, and shows a promising mechanism when combined with multi-temporal observations.

This paper addresses the critical challenge of semantic segmentation for remote sensing images (RSIs) under extremely limited labeled data. A high-quality initial model is paramount for downstream semi-supervised or weakly supervised learning paradigms, as it mitigates error propagation from the outset. We conducted a systematic investigation into self-supervised pretraining to serve this precise need. Within the low-label regime, we identify and tackle two pivotal factors limiting performance: (1) the domain shift between large-scale pretraining data and specific target tasks, and (2) the deficiency in local feature learning caused by large-window masking in visual foundation model (VFM) pretraining. To resolve these issues, we first benchmark various pretraining strategies, demonstrating that a two-phase General-Purpose Pretraining (GPPT) followed by Domain-Adaptive Pretraining (DAPT) framework is optimal, significantly outperforming both single-phase methods and the existing two-phase paradigm initialized from ImageNet. Subsequently, we propose an Edge-Guided Masked Image Modeling (EGMIM) method for the DAPT phase, which explicitly integrates edge priors to guide the masking and reconstruction process, thereby enhancing the model’s capability to capture fine-grained local structures. Extensive experiments on four RSI benchmarks validate the effectiveness of our approach, showing consistent and substantial gains, particularly in extreme low-label scenarios. Beyond empirical results, we provide in-depth mechanistic analyses to explain the synergistic roles of GPPT and DAPT.

Semantic segmentation of remote sensing images is crucial for geospatial applications but is severely hampered by the prohibitive cost of pixel-level annotations. Although semi-supervised learning (SSL) offers a solution by leveraging unlabeled data, its application to Vision Transformers (ViTs) often encounters overfitting and even training instability under extreme label scarcity. To tackle these challenges, we propose a Curriculum-based Self-supervised and Semi-supervised Pipeline (CSSP). The pipeline adopts a staged, easy-to-hard training strategy, commencing with in-domain pretraining for robust feature representation, followed by a carefully designed finetuning stage to prevent overfitting. The pipeline further integrates a novel Difficulty-Adaptive ClassMix (DA-ClassMix) augmentation that dynamically reinforces underperforming categories and a Progressive Intensity Adaptation (PIA) strategy that systematically escalates augmentation strength to maximize model generalization. Extensive evaluations on the Potsdam, Vaihingen, and Inria datasets demonstrate state-of-the-art performance. Notably, with only 1/32 of the labeled data on the Potsdam dataset, the CSSP reaches 82.16% mIoU, nearly matching the fully supervised result (82.24%). Furthermore, we extend the CSSP to a semi-supervised domain adaptation (SSDA) scenario, termed Cross-Domain CSSP (CDCSSP), which outperforms existing SSDA and unsupervised domain adaptation (UDA) methods. This work establishes a stable and highly effective framework for training ViT-based segmentation models with minimal annotation overhead.

In complex agricultural landscapes, especially with limited sample sizes, current instance segmentation techniques often grapple with defining clear boundaries, thereby compromising segmentation accuracy. To mitigate this challenge, our research proposes enhancements through quantitative comparisons and the incorporation of the SAM model to bolster instance segmentation networks. Specifically, we have formulated SAM+Mask2Former and SAM+Mask R-CNN by adopting SAM's feature encoder as the foundational backbone. Further refinement is accomplished by integrating the Efficient Channel Attention (ECA) module and optimizing prompts, culminating in the SAM+ECA+Prompter model. Impressively, these models exhibit robust generalization capabilities when applied to high-resolution remote sensing imagery, even when based on smaller datasets. Empirical evidence demonstrates that, even with a mere 50 samples, the models achieve an Average Precision (AP) and Average Recall (AR) of 0.715. When scaled to 200 samples, AP surges to 0.875, while AR surpasses 0.903—figures that stand toe-to-toe with traditional field segmentation paradigms. Consequently, our research significantly elevates the efficacy of segmentation models, emphasizing a reduction in both labeling and computational expenses, thereby presenting a resource-efficient, high-precision solution for precision agriculture.