Observational data on ocean subsurface temperature and salinity are patently insufficient because in situ observations are complex and costly, while satellite remote-sensed measurements are abundant but mainly focus on sea surface data. To make up for the ocean interior data shortage and entirely use the abundant satellite data, we developed a data-driven deep learning model named Convformer to reconstruct ocean subsurface temperature and salinity fields from satellite-observed sea surface data. Convformer is designed by deeply optimizing Vision Transformer and ConvLSTM, consisting of alternating residual connections between multiple temporal and spatial attention blocks. The input variables consist of sea surface temperature (SST), sea surface salinity (SSS), sea surface height (SSH), and sea surface wind (SSW). Our results demonstrate that Convformer exhibits superior performance in estimating the temperature-salinity structure of the tropical Pacific Ocean. The all-depth average root mean square error (RMSE) of the reconstructed subsurface temperature (ST)/subsurface salinity (SS) is 0.353 °C/0.0695 PSU, with correlation coefficients (R²) of 0.98663/0.99971. In the critical thermocline, although the root mean square errors of ST and SS reach 0.85 °C and 0.121 PSU, respectively, they remain smaller compared to other models. Furthermore, we assessed Convformer’s performance from various perspectives. Notably, we also delved into the potential of Convformer to extract physical and dynamic information from a model mechanism perspective. Our study offers a practical approach to reconstructing the subsurface temperature and salinity fields from satellite-observed sea surface data.

Unsupervised domain adaptation (UDA) aims to improve model performance in the target domain by leveraging labeled data from the source domain while not requiring labeled data in the target domain. It has been widely applied in cross-domain semantic segmentation of remote sensing images (RSI). Despite some advancements in this area, challenges such as class confusion due to color and texture similarities, and class imbalance due to significant scale variations and sample imbalance continue to impede progress in UDA for RSIs segmentation. To address these challenges, we propose a novel self-supervised teacher-student network framework, including two innovative techniques: mask-enhanced class mix (MECM) and scale-based rare class sampling (SRCS). The MECM method applies a high proportion of masks to mixed images derived from both source domain images and target domain images, which encourages the model to infer the semantic information of masked areas from the surrounding context, enhancing cross-domain contextual semantic learning and improving the recognition accuracy of similar classes. Additionally, SRCS increases the sampling proportion of small-scale rare classes, mitigating the issue of class imbalance. Experiments show that our method outperforms existing UDA techniques in terms of PA, mF1, and mIoU, achieving state-of-the-art results on three public datasets. Notably, in the Potsdam IRRG to Vaihingen UDA scenario, our method’s performance on the key metric, mIoU, even surpasses that of supervised training, demonstrating the superiority of our approach. Codes is available at https://github.com/Qiuyb-ai/UDA-With-ME-and-BS.

Road extraction in remote sensing images is crucial for urban planning, traffic navigation, and mapping. However, certain lighting conditions and compositional materials often cause roads to exhibit colors and textures similar to the background, leading to incomplete extraction. Additionally, the elongated and curved road morphology conflicts with the rectangular receptive field of traditional convolution. These challenges significantly affect the accuracy of road extraction in remote sensing images. To address these issues, we propose an end-to-end low-contrast road extraction network called LCMorph, which leverages both frequency cues and morphological perception. First, Frequency-Aware Modules (FAMs) are introduced in the encoder to extract frequency cues, effectively distinguishing low-contrast roads from the background. Subsequently, Morphological Perception Blocks (MPBlocks) are employed in the decoder to adaptively adjust the receptive field to the elongated and curved nature of roads. MPBlock relies on snake convolution, which mimics snakes’ twisting behavior for accurate road extraction. Experiments demonstrate that our method achieves state-of-the-art performance in terms of F1 score and IoU on the self-constructed low-contrast road dataset (LC-Roads), as well as the public DeepGlobe and Massachusetts Roads datasets.

Estimating the intensity of tropical cyclones (TCs) is beneficial for preventing and reducing the impact of natural disasters. Most existing methods for estimating TC intensity utilize single-satellite or single-band remote sensing images, but they lack the ability to quantify the uncertainty of the estimation results. However, TC, as a typical chaotic system, often requires confidence intervals for intensity estimates in real-world emergency decision-making scenarios. Additionally, the use of multi-source image inputs contributes to the uncertainty of the model. Consequently, this study introduces a neural network (MTCIE) that utilizes multi-source satellite images to provide probabilistic estimates of TC intensity. The model utilizes infrared and microwave images from multiple satellites as inputs. It uses a dual-branch self-attention encoder to extract TC image features and provides uncertainty estimates for TC intensity. Furthermore, a dataset for estimating the intensity of multi-source TC remote sensing images (MTCID) is constructed through the registration of latitude, longitude, and time, along with data augmentation. The proposed method achieves a MAE of 7.42 kt in deterministic estimation, comparable to mainstream networks like TCIENet. In uncertain estimation, it outperforms methods like MC Dropout in the PICP metric, providing reliable probability estimates. This supports TC disaster emergency decision making, enhancing risk mitigation in real-world applications.

As a promising paradigm, mobile crowdsensing (MCS) takes advantage of sensing abilities and cooperates with multi-agent reinforcement learning technologies to provide services for users in large sensing areas, such as smart transportation, environment monitoring, etc. In most cases, strategy training for multi-agent reinforcement learning requires substantial interaction with the sensing environment, which results in unaffordable costs. Thus, environment reconstruction via extraction of the causal effect model from past data is an effective way to smoothly accomplish environment monitoring. However, the sensing environment is often so complex that the observable and unobservable data collected are sparse and heterogeneous, affecting the accuracy of the reconstruction. In this paper, we focus on developing a robust multi-agent environment monitoring framework, called self-interested coalitional crowdsensing for multi-agent interactive environment monitoring (SCC-MIE), including environment reconstruction and worker selection. In SCC-MIE, we start from a multi-agent generative adversarial imitation learning framework to introduce a new self-interested coalitional learning strategy, which forges cooperation between a reconstructor and a discriminator to learn the sensing environment together with the hidden confounder while providing interpretability on the results of environment monitoring. Based on this, we utilize the secretary problem to select suitable workers to collect data for accurate environment monitoring in a real-time manner. It is shown that SCC-MIE realizes a significant performance improvement in environment monitoring compared to the existing models.

A major computational bottleneck in classifying large-scale hyperspectral images (HSI) is the mandatory data decompression prior to processing. Compressed-domain computing offers a solution by enabling deep learning on partially compressed data. However, existing compressed-domain methods are predominantly tailored for the Discrete Cosine Transform (DCT) used in natural images, while HSIs are typically compressed using the Discrete Wavelet Transform (DWT). The fundamental structural mismatch between the block-based DCT and the hierarchical DWT sub-bands presents two core challenges: how to extract features from multiple wavelet sub-bands, and how to fuse these features effectively? To address these issues, we propose a novel framework that extracts and fuses features from different DWT sub-bands directly. We design a multi-branch feature extractor with sub-band feature alignment loss that processes functionally different sub-bands in parallel, preserving the independence of each frequency feature. We then employ a sub-band cross-attention mechanism that inverts the typical attention paradigm by using the sparse, high-frequency detail sub-bands as queries to adaptively select and enhance salient features from the dense, information-rich low-frequency sub-bands. This enables a targeted fusion of global context and fine-grained structural information without data reconstruction. Experiments on three benchmark datasets demonstrate that our method achieves classification accuracy comparable to state-of-the-art spatial-domain approaches while eliminating at least 56% of the decompression overhead.