Utilizing a temperature time-series prediction model to achieve good results can help us to accurately sense the changes occurring in temperature levels in advance, which is important for human life. However, the random fluctuations occurring in a temperature time series can reduce the accuracy of the prediction model. Decomposing the time-series data prior to performing a prediction can effectively reduce the influence of random fluctuations in the data and consequently improve the prediction accuracy results. In the present study, we propose a temperature time-series prediction model that combines the seasonal-trend decomposition procedure based on the loess (STL) decomposition method, the jumps upon spectrum and trend (JUST) algorithm, and the bidirectional long short-term memory (Bi-LSTM) network. This model can achieve daily average temperature predictions for cities located in China. Firstly, we decompose the time series into trend, seasonal, and residual components using the JUST and STL algorithms. Then, the components determined by the two methods are combined. Secondly, the three components and original data are fed into the two-layer Bi-LSTM model for training purposes. Finally, the prediction results achieved for both the components and original data are merged by learnable weights and output as the final result. The experimental results show that the average root mean square and average absolute errors of our proposed model on the dataset are 0.2187 and 0.1737, respectively, which are less than the values 4.3997 and 3.3349 attained for the Bi-LSTM model, 2.5343 and 1.9265 for the EMD-LSTM model, and 0.9336 and 0.7066 for the STL-LSTM model.

Land Surface Temperature (LST) is a key parameter in climate systems. The methods for retrieving LST from hyperspectral thermal infrared data either require accurate atmospheric profile data or require thousands of continuous channels. We aim to retrieve LST for natural land surfaces from hyperspectral thermal infrared data using an adapted multi-channel method taking Land Surface Emissivity (LSE) properly into consideration. In the adapted method, LST can be retrieved by a linear function of 36 brightness temperatures at Top of Atmosphere (TOA) using channels where LSE has high values. We evaluated the adapted method using simulation data at nadir and satellite data near nadir. The Root Mean Square Error (RMSE) of the LST retrieved from the simulation data is 0.90 K. Compared with an LST product from the Spinning Enhanced Visible and Infrared Imager (SEVIRI) on Meteosat, the error in the LST retrieved from the Infared Atmospheric Sounding Interferometer (IASI) is approximately 1.6 K. The adapted method can be used for the near-real-time production of an LST product and to provide the physical method to simultaneously retrieve atmospheric profiles, LST, and LSE with a first-guess LST value. The limitations of the adapted method are that it requires the minimum LSE in the spectral interval of 800–950 cm⁻¹ larger than 0.95 and it has not been extended for off-nadir measurements.

As an important component in the surface radiation budget, surface upwelling longwave radiation (SULR) is an outcome of the land surface energy exchange and mainly represents the capability of thermal radiation from the surface of the Earth. Existing satellite-derived SULR products are too coarse to support high-resolution numerical models, and their accuracy needs to be improved. In this study, an equivalent temperature is introduced through which a “split-window” atmospheric correction algorithm is developed for MODIS data to estimate the instantaneous clear-sky SULR. It is a simple and feasible method that is particularly applicable to MODIS data to acquire relatively high precision SULR under clear skies from which qualified water vapor contents (WVC) and thermal channel brightness temperatures are available. The root mean square errors (RMSEs) are less than 13 W/m² for all WVC sub-ranges with the viewing zenith angle (VZA) less than 30°, or for all sub-ranges with the VZA less than 60° and the WVC less than 3.5 g/cm². Also, applications and comparisons with the LST-emissivity method are made by using ground measurements which are collected from the network of surface radiation budget network data (SURFRAD) at the moment of MODIS overpass. Results show that the proposed model has high computational efficiency to estimate SULR from MODIS cloud-free data.

Existing fusion rules focus on retaining detailed information in the source image, but as the thermal radiation information in infrared images is mainly characterized by pixel intensity, these fusion rules are likely to result in reduced saliency of the target in the fused image. To address this problem, we propose an infrared and visible image fusion model based on significant target enhancement, aiming to inject thermal targets from infrared images into visible images to enhance target saliency while retaining important details in visible images. First, the source image is decomposed with multi-level Gaussian curvature filtering to obtain background information with high spatial resolution. Second, the large-scale layers are fused using ResNet50 and maximizing weights based on the average operator to improve detail retention. Finally, the base layers are fused by incorporating a new salient target detection method. The subjective and objective experimental results on TNO and MSRS datasets demonstrate that our method achieves better results compared to other traditional and deep learning-based methods.

Trust prediction is essential to enhancing reliability and reducing risk from the unreliable node, especially for online applications in open network environments. An essential fact in trust prediction is to measure the relation of both the interacting entities accurately. However, most of the existing methods infer the trust relation between interacting entities usually rely on modeling the similarity between nodes on a graph and ignore semantic relation and the influence of negative links (e.g., distrust relation). In this paper, we proposed a relation representation learning via signed graph mutual information maximization (called SGMIM). In SGMIM, we incorporate a translation model and positive point-wise mutual information to enhance the relation representations and adopt Mutual Information Maximization to align the entity and relation semantic spaces. Moreover, we further develop a sign prediction model for making accurate trust predictions. We conduct link sign prediction in trust networks based on learned the relation representation. Extensive experimental results in four real-world datasets on trust prediction task show that SGMIM significantly outperforms state-of-the-art baseline methods.