Unsupervised Domain Adaptation for Remote Sensing Semantic Segmentation (UDA-RSSeg) is to adapt a model trained on the source domain data to the target domain samples, thereby minimizing the need for annotated data across diverse remote sensing scenes. In urban planning and monitoring, the task of UDA-RSSeg on Very-High-Resolution (VHR) images has garnered significant research interest. While recent deep learning techniques have demonstrated huge success in tackling the UDA-RSSeg task for VHR urban scenes, a persistent challenge in addressing the domain shift issue remains. Specifically, there are two primary problems: (1) severe inconsistencies in feature representation across diverse domains, characterized by notably differing data distributions, and (2) the domain gap problem due to the representation bias of the source domain patterns when translating features to predictive logits. To solve these problems, we propose a prompted foundation model based hybrid training joint-optimized network (PFM-JONet) for UDA-RSSeg on VHR urban scene. Our approach integrates the notable “Segment Anything Model” (SAM) as prompted foundation model to leverage its robust generalized representation capabilities, thereby alleviating feature inconsistencies. Based on the feature extracted by SAM-Encoder, we introduce a mapping decoder designed to convert SAM-Encoder features into predictive logits. Additionally, a prompted segmentor is employed to generate class-agnostic maps, which guide the mapping decoder’s feature representations. To efficiently optimize the entire network in an end-to-end manner, we design a hybrid training scheme that integrates feature-level and logits-level adversarial training strategies alongside a self-training mechanism. This scheme enhances the model from diverse, compatible perspectives. To evaluate the performance of our proposed PFM-JONet, we conduct extensive experiments on urban scene benchmark datasets, including ISPRS (Potsdam/Vaihingen) and CITY-OSM (Paris/Chicago). On ISPRS dataset, PFM-JONet surpasses previous SOTA methods by 1.60% in mean IoU value across four adaptation tasks. For CITY-OSM’s adaptation task, it outperforms SOTA by 4.84% in mean IoU value. These results demonstrate the effectiveness of our method. Furthermore, visualization and analysis reinforce the method’s interpretability. The code of this paper is available at https://github.com/CV-ShuchangLyu/PFM-JONet.

The advancement of deep neural network has propelled the widespread application of remote sensing target detection. However, compared to natural scenes, remote sensing targets possess inherent characteristics such as weak features and small scale, leading to a significant performance gap in traditional detection methods. To address these challenges, we undertake a systematic analysis of existing approaches, focusing on two key aspects: inadequate extraction of discriminative features and inappropriate regression measurement metrics. To tackle the first issue, an Instance-level Neighboring Enhancement Network (INEN) is proposed, enhancing the network’s feature extraction capability through inter-object feature aggregation. To address the second issue, a novel metric, Smooth Wasserstein Loss (SWL), is devised. Building upon these principles, a new tiny object detection network for remote sensing images is developed. Extensive experiments on AI-TOD v1/v2 and DOTA v2 remote sensing tiny target detection datasets demonstrate that our approach achieves state-of-the-art (SOTA) performance. Codes are available at https://github.com/sevenwgb/SWIN-TOD.

Pine wilt disease (PWD) is a worldwide affliction that poses a significant menace to forest ecosystems. The swift and precise identification of pine trees under infection holds paramount significance in the proficient administration of this ailment. The progression of remote sensing and deep learning methodologies has propelled the utilization of target detection and recognition techniques reliant on remote sensing imagery, emerging as the prevailing strategy for pinpointing affected trees. Although the existing object detection algorithms have achieved remarkable success, virtually all methods solely rely on a Digital Orthophoto Map (DOM), which is not suitable for diseased trees detection, leading to a large false detection rate in the detection of easily confused targets, such as bare land, houses, brown herbs and so on. In order to improve the ability of detecting diseased trees and preventing the spread of the epidemic, we construct a large-scale PWD detection dataset with both DOM and Digital Surface Model (DSM) images and propose a novel detection framework, DDNet, which makes full use of the spectral features and geomorphological spatial features of remote sensing targets. The experimental results show that the proposed joint network achieves an AP50 2.4% higher than the traditional deep learning network.

Remote sensing scene zero-shot object detection aims to detect and recognize both seen and unseen catrgories of landscape elements with the guidance of the word embeddings. In this task, two primary challenges are identified. Firstly, there exists considerable variability within categories of landscape elements, causing a misalignment between visual features and word embeddings, particularly noticeable for unseen categories. Secondly, existing detection models struggle to provide accurate localization predictions, greatly impacting overall performance. To address these two issues, we propose WEA-DINO (Word Embedding Alignment-DINO). Based on the original DINO structure, our WEA-DINO-Head is specifically designed to align the hidden features of “matching queries” with word embedding features, effectively addressing the misalignment issue between visual features and word embeddings. Furthermore, aligning the hidden features of “denoising queries” with word embedding features enables the translation of localization capabilities from known categories to previously unseen ones. Through extensive experimentation on the DIOR benchmark dataset, our method demonstrates state-of-the-art performance. The code is available at https://github.com/cv516Buaa/WEA-DINO.

Class-incremental semantic segmentation focuses on updating the segmentation model with only new-class samples. Catastrophic forgetting and background shift are the two prevalent challenges. We identify two additional issues in remote sensing data that worsen these problems: significant class distribution variability and error accumulation-induced model degradation. To solve these three problems, we propose a new Self-Training and Curriculum Learning Guided Dynamic Refined Network (STCL-DRNet). First, we introduce a self-training auxiliary branch to complement the frozen last-step model, integrating cross-step knowledge to mitigate rapid forgetting. Then, a gradient-oriented Dynamic Refined Loss is proposed to assess under-learned classes and mitigate class imbalance. Furthermore, class-balanced curriculum learning is embedded to alleviate performance degradation throughout incremental training. Extensive experiments on benchmark datasets, including DeepGlobe, iSAID, ISPRS Potsdam, and Vaihingen, demonstrate that the proposed STCL-DRNet achieves state-of-the-art (SOTA) performance. In the 1-1s setting of the DeepGlobe dataset, STCL-DRNet exceeds previous SOTA methods by 11.6% in mIoU. For the iSAID 10-1s setting, it outperforms the previous SOTA by 12.76% in mIoU. As for ISPRS Potsdam and Vaihingen, our STCL-DRNet surpasses the SOTA by 5%-8% in all settings. Visualization and analysis further validate its interpretability. Our code is available at https://github.com/cv516Buaa/STCL-DRNet.

Deterministic sea wave reconstruction techniques are critical for enhancing maritime safety and disaster warnings. Coherent radar remote sensing captures sea surface velocity information to enable more precise wave reconstruction. Existing difference matrix methods address rank-deficient systems through artificial boundary processing, which distorts local hydrodynamic characteristics and propagates errors to global features, thereby limiting the accuracy and stability of reconstructions. To resolve this limitation, this study proposes a physics-constrained deterministic wave reconstruction methodology. We introduce the Data-Anchored Projection model for the differential matrix, extracting hydrodynamic constraints directly from radar backscatter data. This approach achieves stable solutions for rank-deficient systems without artificial boundaries. The model’s performance was rigorously validated through both simulated and real-sea experiments. The simulation results demonstrate a minimum 13% accuracy improvement over conventional methods and high stability under various sea states and at different range resolutions. In a real-sea trial under sea states 3 to 5, reconstruction errors remained below 10%, with consistent stability observed across varying sea states.