Automatically classifying a huge amount of ground‐based auroral images is essential to facilitate aurora morphology statistical research and aid in comprehending the magnetospheric dynamics. However, facing the challenge of insufficient labeled images, deep learning methods perform suboptimally on small auroral image data sets, and traditional machine learning methods based on handcrafted features heavily rely on expert knowledge. In this paper, we propose a novel method that leverages the merits of both traditional machine learning and deep learning methods by extracting deep second‐order tensor features to train a support vector machine (SVM). To improve compactness and discriminative ability of the features, we comply the intrinsic data geometry on Riemannian manifold to employ dimensionality reduction and map the dimensionality‐reduced features from Riemannian space to Euclidean space for the SVM classifier. Experimental results on small aurora data sets conclusively demonstrate the effectiveness of our method, exhibiting competitive performance with recent aurora classification methods.

Sea fog is a natural phenomenon that significantly reduces visibility, posing navigational hazards for ships and impacting coastal activities. Geostationary meteorological satellite data have proven to be indispensable for sea fog monitoring due to their large spatial coverage and spatiotemporal consistency. However, the spectral similarities between sea fog and low clouds result in omissions and misclassifications. Furthermore, high clouds obscure certain sea fog regions, leading to under-detection and high false alarm rates. In this paper, we present a novel sea fog detection method to alleviate the challenges. Specifically, the approach leverages a fusion of spectral, motion, and spatiotemporal texture consistency features to effectively differentiate sea fog and low clouds. Additionally, a multi-scale self-attention module is incorporated to recover the sea fog region obscured by clouds. Based on the spatial distribution characteristics of sea fog and clouds, we redesigned the loss function to integrate total variation loss, focal loss, and dice loss. Experimental results validate the effectiveness of the proposed method, and the detection accuracy is compared with the vertical feature mask produced by the CALIOP and exhibits a high level of consistency.

Accurate classification of ground-based auroral images is essential for studying variations in auroral morphology and uncovering magnetospheric mechanisms. However, distinguishing subtle morphological differences among different categories of auroral images presents a significant challenge. To excavate more discriminative information from ground-based auroral images, a novel method named learning representative channel attention information from second-order statistics (LRCAISS) is proposed. The LRCAISS is highlighted with two innovative techniques: a second-order convolutional network and a novel second-order channel attention block. The LRCAISS extends from Resnet50 architecture by incorporating a second-order convolutional network to capture more detailed statistical representation. Meanwhile, the novel second-order channel attention block effectively recalibrates these features. LACAISS is evaluated on two public ground-based auroral image datasets, and the experimental results demonstrate that LRCAISS achieves competitive performance compared to existing methods.

Accurate and timely reconstruction of sea surface wind fields from available scatterometer‐derived observations is essential for rapid‐response forecasting and operational oceanography. In this study, we explore a novel physics‐guided generative learning network to reconstruct sea surface wind vector fields directly from sparse scatterometer‐like inputs. The proposed method integrates data‐driven learning with physics‐based guiding. The discriminator is designed to assess the physical plausibility of reconstructed wind fields through physics consistency scores, while the generator is guided by a composite loss function incorporating tailored physical terms to ensure physically faithful and dynamically coherent outputs. Experimental results show that the proposed method achieves fast, structurally preservative, and physically consistent reconstructions, even under significant data sparsity. Notably, during Typhoon Merbok (September 2022), the generative learning network effectively recovered the vortex structure with a wind speed RMSE no more than 2 m/s, demonstrating strong capability in handling extreme events.

Cloud cover estimation is of crucial significance in meteorological observations and short-term/long-term weather forecasting, as it directly affects the accuracy of radiation balance assessment, precipitation prediction, and climate change modeling. Ground-based automated cloud quantification observation instruments enable continuous, high-resolution cloud monitoring with spatial-temporal continuity that satellite remote sensing cannot fully achieve, highlighting the immense value of ground-based cloud image processing for practical meteorological applications. However, existing cloud detection methods predominantly rely on supervised training with ground truth masks, which overlook the rich contextual information and inherent regularization constraints embedded in original cloud images. This oversight frequently results in mismatched cloud boundaries, inadequate model interpretability, and poor adaptability to complex cloud morphologies—particularly for thin clouds and cirrus clouds characterized by weak grayscale contrast, sparse texture, and irregular shapes. Consequently, these limitations lead to suboptimal detection performance, including under-segmentation or over-segmentation, and further induce inaccuracies in quantitative cloud cover estimation.To address the aforementioned issues and achieve accurate cloud cover detection results, this study proposes a model-agnostic refinement method designed to optimize the coarse detection masks generated by any pre-trained cloud detection model. The framework is jointly optimized by three loss functions: a local similarity descriptor, total variation (TV) regularization, and a traditional detection loss (e.g., cross-entropy). Specifically, the local similarity descriptor is defined as the difference between two terms: the average grayscale difference of each pixel and cloud region and background pixels within a local window. This descriptor effectively enhances the discriminability between cloud and non-cloud regions at the local level. The total variation regularization term is introduced to maintain the smoothness of the detection boundary and suppress spurious noise. The cross-entropy loss ensures the overall consistency between the refined result and the ground truth.Minimizing the combined loss function drives the coarse detection result to evolve adaptively along the actual cloud boundary, thereby achieving more precise alignment with the true cloud contours. Notably, the proposed framework elevates the detection of thin clouds and cirrus clouds, effectively mitigating missed detection areas in these tenuous cloud structures. Furthermore, the integrated loss function enhances model interpretability: the local similarity descriptor explicitly quantifies the differences within local window, and minimizing this term inherently refines the detection by strengthening the distinction between cloud and background regions. Ultimately, the refined detection results substantially improve the accuracy of cloud cover estimation, laying a solid foundation for reliable meteorological observations and weather forecasting applications.