The El Niño-Southern Oscillation (ENSO) event often causes natural disasters in mainland China. Existing quantitative analysis of ENSO event’s effects on climate change in mainland China is insufficient. The monthly scale prediction effectiveness of ENSO events is still low. Global Navigation Satellite System (GNSS) can estimate zenith tropospheric delay (ZTD) with high accuracy, which can study ZTD responses to ENSO and improve the prediction accuracy of ENSO events. This study quantitatively analyzed the response patterns of GNSS ZTD time–frequency variation to ENSO events in mainland China. The monthly multivariate ENSO index (MEI) thresholds for GNSS ZTD anomaly response to ENSO events are (−1.12, 1.92) for the tropical monsoon zone (TPMZ), (−1.12, 1.61) for the subtropical monsoon zone (SMZ), (−1.19, 1.62) for the temperate monsoon zone (TMZ), (−1.26, 1.64) for the temperate continental zone (TCZ), and (−1.22, 1.72) for the mountain plateau zone (MPZ). The ENSO event causes the amplitude of the nine-month variation period to decrease and the amplitude of the 0.8–3-month period to increase for the GNSS ZTD in mainland China. Furthermore, a forecasting model is proposed by integrating fast Fourier transform and long short-term memory extended (FFT-LSTME). The model uses monthly MEI as the primary input and the GNSS ZTD reconstruction sequence that responds to ENSO as the auxiliary input. It can predict ENSO events in the next 24 months with an index of agreement (IA) of 91.56% and a root mean square error (RMSE) of 0.25. The RMSE is optimized by 70.48%, 43.95%, and 11.6% when compared with radial basis function (RBF), LSTM, and FFT-LSTM.

Multi-Object Tracking (MOT) mainly implements certain complex multi-step tracking by using detection algorithms, which respectively perform target detection, feature extraction and data association. The main challenges of Unmanned Aerial Vehicle (UAV) tracking are complex background and mutual occlusion. In this work, we propose a simple online UAV tracking with Transformer. It uses the encoder-decoder mechanism to introduce a set of object query in the pipeline, which can achieve the detection of new targets accurately. This method uses an online joint-detection-and-tracking pipeline based on the encoder-decoder mechanism. The complex and multi-step components in the previous method are simplified. Further, it is a new structure based on Transformer. Object query detects the object in the current frame. The object feature query in the previous frame associates these current objects with the previous object. We proposed a method that can be well applied to small targets such as UAVs, achieved 64.1% MOTA competition on the MOT15 challenge dataset.

The Kernel Correlation Filter Tracking Algorithm(KCF)is a lightweight tracking algorithm with the advantages of fast tracking speed and good effect. However, when the target is occluded and the scale changes, the algorithm will have tracking drift and tracking loss. Aiming at the kernel-related filter tracking algorithm that cannot solve the tracking failure caused by occlusion, target scale changes and other factors in the tracking process, an anti-occlusion and scale-adaptive kernel-related filtering algorithm is proposed. We build a scale pool and use the scale pool to train a one-dimensional fast scale filter to solve the problem of target scale changes. This paper uses the average occlusion distance metric and the size of the context occlusion factor to determine the occlusion state of the target, and dynamically select the learning update rate of the target model according to the target occlusion state. When it is judged that the target is severely occluded, the target position is predicted according to the previous motion state of the target, and the small-range re-detection positioning mechanism proposed in this paper is used to re-detect the target within a certain range of the predicted position. At the same time, the re-detected target is occluded again Judgment to determine whether the target is out of the occlusion. If it is determined that the target is still severely occluded, it means that the target is not out of the occlusion area, the re-detection of the target position is inaccurate, and the predicted position is output. Experimental results show that the accuracy and success rate of the algorithm in this paper are 0.819 and 0.669, which are 8.33% and 7.04% higher than the KCF algorithm. The tracking effect of this algorithm is better than that of KCF algorithm.

As an important information fusion method in video surveillance forensics, infrared and visible image fusion has received extensive attention and research. In order to reduce the sampling number of the original video surveillance image information in the fusion process, improve the processing efficiency and the background quality of the fused image, a frequency domain fusion algorithm of infrared and visible image based on compressed sensing for video surveillance forensics is proposed. Firstly, the compressed sensing is used for sampling, and then the subspace pursuit (SP) algorithm is applied to reconstruct the sparse coefficient. Secondly, the sparse coefficient is transformed in frequency domain, and the fusion is completed in frequency domain, and finally obtains the fused image. A large number of experiments show when the number of compressed sensing samples is 70%, the fusion quality is better than the other four comparison algorithms that take advantage of full sampling. Compared with similar algorithms, this algorithm has obvious advantages in the presentation of image background information and gradient structure information.