Search

High-resolution imagery and deep learning models have gained increasing importance in land-use mapping. In recent years, several new deep learning network modeling methods have surfaced. However, there has been a lack of a clear understanding of the performance of these models. In this study, we applied four well-established and robust deep learning models (FCN-8s, SegNet, U-Net, and Swin-UNet) to an open benchmark high-resolution remote sensing dataset to compare their performance in land-use mapping. The results indicate that FCN-8s, SegNet, U-Net, and Swin-UNet achieved overall accuracies of 80.73%, 89.86%, 91.90%, and 96.01%, respectively, on the test set. Furthermore, we assessed the generalization ability of these models using two measures: intersection of union and F1 score, which highlight Swin-UNet's superior robustness compared to the other three models. In summary, our study provides a systematic analysis of the classification differences among these four deep learning models through experiments. It serves as a valuable reference for selecting models in future research, particularly in scenarios such as land-use mapping, urban functional area recognition, and natural resource management. [ABSTRACT FROM AUTHOR]

Global climate change, expected to be one of the most severe challenges that human beings have ever encountered, has had far-reaching impacts on ecosystems and humans, among which the potentially increasing chance of violent conflict has raised attention recently. However, several years of research have produced no consensus regarding whether climate variability affects the risk of armed conflict and how it may affect conflict. In this study, we built a geographically disaggregated method to explore the relationship between climate variability from normal climate conditions and armed conflicts both on a local and regional scale. With the 10,993 conflict records acquired in 25 African countries over 16 years from 2000 to 2015, we estimated the effects of temperature and wet day variability on conflicts in agricultural and non-agricultural areas, respectively, on gridded 1° resolution. The results showed that deviations from the normal climate have a systematical impact on the risk of conflict: The risk of violence rises with increasing deviations from the temperature norms in both non-agricultural and agricultural areas. Regarding the rainfall variability, in non-agricultural areas, the risk of violence grows with increasing anomalous wet days, either more or fewer days than the annual average, while in agricultural areas, increases in violence risk only exhibit under the impact of fewer wet days than the annual average. We expect these findings would provide empirical support for policymakers and relevant organizations who need to prepare additional law enforcement and/or peacekeeping resources when climatic anomalies are detected. [ABSTRACT FROM AUTHOR]

Background: Human cystic and alveolar echinococcosis are neglected tropical diseases that WHO has prioritized for control in recent years. Both diseases impose substantial burdens on public health and the socio-economy in China. In this study, which is based on the national echinococcosis survey from 2012 to 2016, we aim to describe the spatial prevalence and demographic characteristics of cystic and alveolar echinococcosis infections in humans and assess the impact of environmental, biological and social factors on both types of the disease. Methods: We computed the sex-, age group-, occupation- and education level-specific prevalences of cystic and alveolar echinococcosis at national and sub-national levels. We mapped the geographical distribution of echinococcosis prevalence at the province, city and county levels. Finally, by analyzing the county-level echinococcosis cases combined with a range of associated environmental, biological and social factors, we identified and quantified the potential risk factors for echinococcosis using a generalized linear model. Results: A total of 1,150,723 residents were selected and included in the national echinococcosis survey between 2012 and 2016, of whom 4161 and 1055 tested positive for cystic and alveolar echinococcosis, respectively. Female gender, older age, occupation at herdsman, occupation as religious worker and illiteracy were identified as risk factors for both types of echinococcosis. The prevalence of echinococcosis was found to vary geographically, with areas of high endemicity observed in the Tibetan Plateau region. Cystic echinococcosis prevalence was positively correlated with cattle density, cattle prevalence, dog density, dog prevalence, number of livestock slaughtered, elevation and grass area, and negatively associated with temperature and gross domestic product (GDP). Alveolar echinococcosis prevalence was positively correlated with precipitation, level of awareness, elevation, rodent density and rodent prevalence, and negatively correlated with forest area, temperature and GDP. Our results also implied that drinking water sources are significantly associated with both diseases. Conclusions: The results of this study provide a comprehensive understanding of geographical patterns, demographic characteristics and risk factors of cystic and alveolar echinococcosis in China. This important information will contribute towards developing targeted prevention measures and controlling diseases from the public health perspective. [ABSTRACT FROM AUTHOR]