Cloud contamination is a common issue that severely reduces the quality of optical satellite images in remote sensing fields. With the rapid development of deep learning technology, cloud contamination is expected to be addressed. In this paper, we propose Denoising Diffusion Probabilistic Model-Cloud Removal (DDPM-CR), a novel cloud removal network that can effectively remove both thin and thick clouds in optical image scenes. Our network leverages the denoising diffusion probabilistic model (DDPM) architecture to integrate both clouded optical and auxiliary SAR images as input to extract DDPM features, providing significant information for missing information retrieval. Additionally, we propose a cloud removal head adopting the DDPM features with an attention mechanism at multiple scales to remove clouds. To achieve better network performance, we propose a cloud-oriented loss that considers both high- and low-frequency image information as well as cloud regions in the training procedure. Our ablation and comparative experiments demonstrate that the DDPM-CR network outperforms other methods under various cloud conditions, achieving better visual effects and accuracy metrics (MAE = 0.0229, RMSE = 0.0268, PSNR = 31.7712, and SSIM = 0.9033). These results suggest that the DDPM-CR network is a promising solution for retrieving missing information in either thin or thick cloud-covered regions, especially when using auxiliary information such as SAR data.

The rapid development of Unmanned Aerial Vehicle (UAV) remote sensing conforms to the increasing demand for the low-altitude very high resolution (VHR) image data. However, high processing speed of massive UAV data has become an indispensable prerequisite for its applications in various industry sectors. In this paper, we developed an effective and efficient seam elimination approach for UAV images based on Wallis dodging and Gaussian distance weight enhancement (WD-GDWE). The method encompasses two major steps: first, Wallis dodging was introduced to adjust the difference of brightness between the two matched images, and the parameters in the algorithm were derived in this study. Second, a Gaussian distance weight distribution method was proposed to fuse the two matched images in the overlap region based on the theory of the First Law of Geography, which can share the partial dislocation in the seam to the whole overlap region with an effect of smooth transition. This method was validated at a study site located in Hanwang (Sichuan, China) which was a seriously damaged area in the 12 May 2008 enchuan Earthquake. Then, a performance comparison between WD-GDWE and the other five classical seam elimination algorithms in the aspect of efficiency and effectiveness was conducted. Results showed that WD-GDWE is not only efficient, but also has a satisfactory effectiveness. This method is promising in advancing the applications in UAV industry especially in emergency situations.

The examination of poverty-causing factors and their mechanisms of action in poverty-stricken villages is an important topic associated with poverty reduction issues. Although the individual or background effects of multilevel influencing factors have been considered in some previous studies, the spatial effects of these factors are rarely involved. By considering nested geographic and administrative features and integrating the detection of individual, background, and spatial effects, a bilevel hierarchical spatial linear model (HSLM) is established in this study to identify the multilevel significant factors that cause poverty in poor villages, as well as the mechanisms through which these factors contribute to poverty at both the village and county levels. An experimental test in the region of the Wuling Mountains in central China revealed the following findings. (1) There were significant background and spatial effects in the study area. Moreover, 48.28% of the overall difference in poverty incidence in poor villages resulted from individual effects at the village level. Additionally, 51.72% of the overall difference resulted from background effects at the county level. (2) Poverty-causing factors were observed at different levels, and these factors featured different action mechanisms. Village-level factors accounted for 14.29% of the overall difference in poverty incidence, and there were five significant village-level factors. (3) The hierarchical spatial regression model was found to be superior to the hierarchical linear model in terms of goodness of fit. This study offers technical support and policy guidance for village-level regional development.

Emergency response mechanisms were activated throughout China during the COVID-19 outbreak. It is different from the temporary, partial, and limited pollution control measures taken to ensure the regional environmental quality during several important events such as the 2008 Beijing Olympic Games and the 2014 Asia-Pacific Economic Cooperation (APEC). During the COVID-19 epidemic period, extensive movement of people and almost all unnecessary industrial production (necessary industrial production refers to the production of food, epidemic prevention materials, etc.) have been severely restricted, so transportation and industrial production have been greatly reduced. This is a rare extreme emission reduction scenario that presents a unique opportunity for atmospheric research. In this study, based on hourly mass concentration data of NO₂ and SO₂ from atmospheric monitoring sites in the Beijing-Tianjin-Hebei (BTH) region during the COVID-19 epidemic period, the changes in transportation and industrial production in the region, data statistics, and spatial analysis were used to analyze the pollution changes and their causes. The results indicate that the NO₂ and SO₂ concentrations in the BTH region decreased significantly during the epidemic period. The spatial distribution pattern of NO₂ pollution in the BTH region was “high in the southeast and low in the northwest,” and SO₂ pollution in the BTH region was high in the southern and eastern parts of Hebei. The initiation of emergency response level 1 had an obvious effect on reducing NO₂ and SO₂ pollution in the region, while the impact of emergency response level 2 and below was limited. Compared with the single traffic control, the comprehensive control, similar to the emergency response, had a better effect on reducing NO₂ pollution in the region. The control of major large cities in the region also had a certain effect on alleviating NO₂ and SO₂ pollution in the entire region. Moreover, for activities under short-term control, it is particularly important to guard against the “retaliatory growth” after the control is lifted. By reducing and controlling some polluting industries in industrial production, the degree of NO₂ and SO₂ pollution in the region can be effectively reduced. The manufacturing industry of chemical raw materials and the chemical products and non-metallic mineral products industry made a great contribution to the change in industrial source pollution emissions in the BTH region during the COVID-19 epidemic. Road traffic emissions remained an important source of NO₂ emissions in the BTH region during this period. NO₂ emission reduction can be effectively achieved by controlling road traffic and transportation.

Most of the traditional image feature point extraction and matching methods are based on a series of light properties of images. These light properties easily conflict with the distinguishability of the image features. The traditional light imaging methods focus only on a fixed depth of the target scene, and subjects at other depths are often easily blurred. This makes the traditional image feature point extraction and matching methods suffer from a low accuracy and a poor robustness. Therefore, in this paper, a light field camera is used as a sensor to acquire image data and to generate a full-focus image with the help of the rich depth information inherent in the original image of the light field. The traditional ORB feature point extraction and matching algorithm is enhanced with the goal of improving the number and accuracy of the feature point extraction for the light field full-focus images. The results show that the improved ORB algorithm extracts not only most of the features in the target scene but also covers the edge part of the image to a greater extent and produces extracted feature points which are evenly distributed for the light field full-focus image. Moreover, the extracted feature points are not repeated in a large number in a certain part of the image, eliminating the aggregation phenomenon that exists in traditional ORB algorithms.