Accurate electroencephalogram(EEG) annotation is essential for seizure detection but costly and error-prone, which can affect subsequent tasks. Moreover, the brain is a non-Euclidean topological structure, which contains spatial information for seizure detection. Based on these, this paper proposes a semi- supervised model based on Gaussian mixture variational autoencoder with graph representation, named GGMVAE. Firstly, for unlabeled EEG signals, we construct an adjacency matrix via Pearson correlation between channels. Then, matrix and EEG features are fed into a Gaussian Mixture VAE to learn the temporal and spatiall features through unsupervised training. Finally, the pre-trained encoder extracts low-dimensional features from partially labeled data for classification. The method is evaluated on the epilepsy dataset at the University of Helsinki, achieving the accuracy of 97.27 %, precision of 96.17 %, recall of 98.41 %, and F1-score of 97.28 %. The results indicate that this semi-supervised model can effectively learn the temporal and spatial features of EEG signals, improving the performance of seizure detection..

Conventional pixel-level seamline detection algorithms exhibit exponential time complexity on large, batch-mode remote-sensing mosaics, making it difficult to achieve an optimal trade-off between accuracy and efficiency. This paper introduces a globally optimal and highly efficient seamline detection framework. First, a preliminary seamline network is generated by iteratively clipping valid orthoimage regions with a Voronoi diagram, and image blocks are extracted only within overlap areas to markedly reduce data volume. Second, a cost graph constructed on down-sampled blocks is traversed in a reverse-diagonal Z-pattern; a “local entropy–gradient” composite cost function is applied, and a linear-time dynamic-programming (DP) scheme rapidly produces coarse seamlines that bypass texture-rich regions and confine the search space to a narrow band. Third, a buffer centered on the coarse seamline is created, within which an enhanced Dijkstra algorithm performs pixel-level refinement to accurately avoid complex obstacles. Experiments on the GF-7 data set demonstrate that, compared with five representative methods—SMP-DP, A*, Dijkstra, graph-cut, and OrthoVista—the proposed approach improves geometric accuracy by 14.46%, 58.69%, 50.20%, 17.79%, and 69.30%, respectively; processing efficiency is increased by 12.74%, 19.19%, 49.89%, >500%, and 83.72%, respectively. The algorithm has successfully mosaicked 627 GF-7 scenes covering the entire Henan Province, and has yielded similarly favorable results on ZY-3, GF-1 and GF-3 imagery, underscoring its high applicability and robustness for multi-source, large-format remote-sensing production.

Nighttime light (NTL) remote sensing data has been widely used in various fields, such as human activity analysis, urbanization studies, and economic evaluation. However, Earth’s nighttime environment is very complex so that the NTL images are seriously affected by atmospheric effect and moonlight. This complexity primarily stems from the numerous atmospheric scattering and absorption, as well as the incoming moonlight, which can significantly distort and contaminate nighttime light observed by the satellite and consequently reduce the precision and stability of NTL data. In order to improve the quantitatively quality of the NTL data, this paper proposes an innovative atmospheric correction algorithm that leverages the nighttime radiative transfer model (nRTM) considering both atmospheric effect and moonlight effect to get ground radiance of artificial lights from satellite nighttime light images. This model takes into account the complex interactions between light and the atmosphere. By simulating these processes, the algorithm is able to separate the contributions of atmospheric scattering and absorption from the original NTL images. To demonstrate the effectiveness of the proposed algorithm, this paper takes the SDGSAT-1 NTL image of Beijing as a representative case study of atmospheric correction. By comparing the corrected and uncorrected images, it is evident that the atmospheric correction significantly improves the quality of the NTL data and the ground nighttime lighting information becomes clearer and more accurate, effectively removing noise interference and enhancing data reliability. Moreover, it also found that the high-pressure sodium (HPS) lamps and LED lamps in the NTL images presented different radiance values and spectral shapes that can be helpful for classifying different lamps.

Night-time light remote sensing has been widely used in various fields, including human activity analysis, urbanization studies, and economic research. However, Earth’s nighttime environment is very complex so that nighttime light images are seriously affected by atmospheric elements and moonlight. To address this issue, this paper proposed an atmospheric correction algorithm on basis of a newly developed radiative transfer model (RTM) that aims to derive the ground radiance of artificial lights from satellite nighttime light images. The SDGSAT-1 night-time light image of Beijing is used as a case study to demonstrate the effectiveness of the proposed algorithm. Results show that the new algorithm can effectively removes the atmospheric effects and improves the data quality of nighttime light images.

The Urban Canopy Multiple-scattering thermal Radiative Transfer (UCM-RT) model, incorporating the effects of urban geometry and adjacent thermal radiation from neighboring pixels, depicts the process of thermal radiation transfer on the urban surface, and therefore provided new opportunity to develop new retrieval algorithms for urban land surface temperature (ULST). This paper aims at developing an urban split-window (USW) algorithm for deriving ULST from high-spatial-resolution thermal infrared (TIR) data from the Visible and Infrared Multispectral Sensor (VIMS) onboard Chinese GaoFen-5 (GF-5) satellite. The VIMS provides 4-channel TIR image with a spatial resolution of 40 m. The coefficients of the USW algorithm were obtained based on several subranges of atmospheric column water vapors (CWV), emissivity and sky view factors (SVFs) under various land surface conditions, by removing the geometry, adjacent and atmospheric effects. Methods of estimating urban pixel emissivity and CWV in urban areas were also conducted. The sensitive analysis of instrument noise and uncertainty of CWV, pixel emissivity and SVFs demonstrated the reliability of the USW algorithm in ULST retrieval. The accuracy evaluation shows that the root-mean-square errors of the ULST results is less than 0.7 K in theory. Compared with the conventional SW algorithms and publicly released LST products, the USW algorithm obtained better results in estimating ULST, especially in high-density building areas. Finally, the USW algorithm is expected to be beneficial to the application of multiple thermal infrared sensor data, for example, the newly launched GF-5 No.2 satellite images.

Thermal infrared (TIR) remote sensing observation signal is influenced by both atmospheric and land surface conditions that are difficult to separate with conventional multichannel TIR data. Because of the advantage of channel wealth, hyperspectral TIR data can simultaneously estimate the land surface and atmospheric parameters using neural network models or integrating them with physical models. However, the commonly used neural network models do not fully explore the correlation between different channels by treating the input data as discrete features. Thus, this study aims to develop a new deep neural network (DNN) by combining the long short-term memory (LSTM) network and convolutional neural network (CNN) for estimating land surface temperature (LST), emissivity, atmospheric transmittance, upward radiance, and downward radiance more accurately. By applying on the thermal airborne hyperspectral imager (TASI) simulation dataset covering global atmospheric conditions with 32 channels in$8.0- 11.5\,\,\mu \text{m}$, the proposed model achieved results with the LST error of 0.95 K, the emissivity error of less than 0.012 for each channel, and the accuracy of three atmospheric parameters has also been improved compared with the current neural network models. Our model has been applied to a real TASI image, and its validity was further proved by the ground measurement validation data. Therefore, it can provide more reliable initial values for physical optimization models.

In the field of remote sensing, due to memory consumption and computational burden, the single-image super-resolution (SISR) methods based on deep convolution neural networks (CNNs) are limited in practical application. To address this problem, we propose a lightweight feature enhancement network (FeNet) for accurate remote-sensing image super-resolution (SR). Considering the existence of equipment with extremely poor hardware facilities, we further design a lighter FeNet-baseline with about 158K parameters. Specifically, inspired by lattice structure, we construct a lightweight lattice block (LLB) as a nonlinear feature extraction function to improve the expression ability. Here, channel separation operation makes the upper and lower branches of the LLB only responsible for half of the features, and the weight coefficients calculated through the attention mechanism enable the upper and lower branches to communicate efficiently. Based on LLB, the feature enhancement block (FEB) is designed in a nested manner to obtain expressive features, where different layers are responsible for the features with different texture richness, and then features from different layers are sequentially fused from deep to shallow. Model parameters and multi-adds operations are used to evaluate network complexity, and extensive experiments on two remote-sensing and four SR benchmark test datasets show that our methods can achieve a good tradeoff between complexity and performance. Our code will be available athttps://github.com/wangzheyuan-666/FeNet.