To accurately capture the complex relationship between PM2.5 and predictive factors, and to obtain spatially continuous PM2.5 pollution distribution with higher resolution and prediction accuracy, a regional PM2.5 pollution early warning mechanism is constructed.In this study, the firefly algorithm-support vector machine (FA-SVM) is used to optimize the land use regression (LUR) model, estimating the PM2.5 mass concentrations in the Fenwei Plain in 2019 at a spatial resolution of 1 km.The results indicate that, compared to conventional LUR and SVM models, FA-SVM demonstrates superior predictive performance.The ten-fold cross-validation coefficient of determination for FA-SVM is as high as 0.90, with a root mean square error and mean absolute error of 12.29 μg/m3 and 8.99 μg/m3, respectively.In contrast, the validation coefficient of determination for LUR and SVM are 0.75 and 0.85, respectively, with root mean square error values of 19.57 μg/m3 and 14.37 μg/m3, and mean absolute error values of 14.84 μg/m3 and 9.62 μg/m3, respectively.PM2.5 pollution in the Fenwei Plain in 2019 exhibits significant spatiotemporal heterogeneity.Temporally, PM2.5 pollution is most severe in winter, gradually decreasing in spring, autumn, and summer.Spatially, areas with relatively higher economic levels show higher PM2.5 mass concentration, forming high-value aggregation zones, while the Qinling Mountains region represents low-value aggregation zones.Overall, PM2.5 exhibits a spatial pattern of higher concentrations in the central region and lower concentrations in the surrounding areas.
In response to the motion blur and visual interference problems resulting from the high-speed movement of unmanned aerial vehicles(UAVs),this paper presents a quality optimi-zation method based on dynamic interpolation and adaptive image enhancement.First,a dynamic blur kernel is constructed in accordance with the target trajectory and image feature point track-ing.By integrating interpolation techniques,a dynamic image interpolation method is developed to restore the crucial details of the image.Subsequently,by incorporating three image enhance-ment techniques such as histogram equalization,an adaptive image re-processing strategy is pro-posed to comprehensively optimize the image quality.Finally,this paper simulates the image deg-radation scenarios of UAVs,creates a corresponding dataset,and conducts comparisons of image quality improvement and verification of the enhancement of its application effects.The experi-mental results indicate that,compared with traditional de-blurring techniques like Wiener filte-ring,the method proposed in this paper minimally improves three metrics,namely the peak sig-nal-to-noise ratio,by 19.3%,55.1%,and 8.7%respectively.In applications based on the YOLO model,it enhances the recognition accuracy by more than 13.4%,thereby strengthening the target identification ability of high-speed UAVs.
User access behavior changes over time and in the environment,making it difficult to accurately capture data features and affecting the effectiveness of user access data mining.There-fore,a new user access data mining method was proposed in the article.Firstly,Z-score and mean imputation methods were used to process user access data.Secondly,for the processed user access data,a relative positional self-attention mechanism was used to capture the relative positional re-lationship,mine the connections between these scattered related access behaviors in a long se-quence,and complete the feature extraction of user access data.Finally,the extracted features were used as clustering centroids to complete user access data mining through K-means cluste-ring.The experimental results show that the method proposed in this paper performs well in data mining accuracy,with normalized mutual information(NMI)values and adjusted rand index(ARI)values both exceeding 0.95,and area under curve(AUC)values close to 1.At the same time,this method also has significant advantages in terms of memory occupancy,with a memory occupancy rate of less than 2%,effectively reducing memory overhead and improving resource u-tilization efficiency.This method has broad application prospects in fields such as user behavior a-nalysis and personalized recommendation.
Aiming at the issue of the shape of Han women's wedding dresses in the Qing dynasty,The existing portraits of women's dress in the Qing dynasty and the old wedding photos retained from the late Qing dynasty were made as research objects in this study.By integrating textual research with visual analysis,an in-depth exploration of the evolution of Han women's wedding dress shapes were conducted during the Qing dynasty.The research shows that the Han women's wedding dress is usually a combina-tion of phoenix coronet,jacket,horse-face skirts and small shoes.In the middle and late Qing dynasty,the cloud shoulder gradually became popular with the embrroidered cape combination,the phoenix coro-net shape evolved into a semi-crown type,and the embrroidered cape evolved into a"vest-like".The jacket gradually integrated into the Manchu costume elements,and the horse-face skirts evolved into skirts such as pleated skirts,Langan skirts,phoenix-tailed skirts,and Yuehua skirts.The research con-clusion has reference significance for the study of traditional wedding dress shape in China.
Solar photocatalytic natural ventilation has attracted wide attention because of its pas-sive ventilation and indoor pollutant purification effect,but it has low ventilation and purification efficiency.Using mechanical ventilation to assist in the construction of photocatalytic multiple ventilation can effectively overcome the lack of passive ventilation,and improve the purification efficiency of indoor pollutants.A pollutant flow diffusion model coupled between photocatalytic multiple ventilation wall and indoor environment was constructed.Secondly,the connection be-tween the structural optimization of the ventilation wall and the indoor air flow distribution was analyzed.Finally,the influence of mechanical ventilation assistance on indoor pollutant ventilation and purification energy efficiency of the model system was analyzed to verify the practicability and indoor comfort of the model.The results show that the average indoor concentration of the me-chanically assisted solar photocatalytic natural ventilation model can be reduced by 1 step com-pared with the traditional model,and the purification rate can be up to 29％.When the inclination height of the model is 0.15 m,the ventilation performance is optimal and the ventilation volume can reach 0.349 m3/s.The thermal performance of the system model is optimized and can be re-duced by up to 11 J,which can effectively avoid the indoor overheating problem of the traditional ventilation wall model in summer.The research fully considers the optimization of building ener-gy saving and indoor comfort and reduces the seasonal limitation impact of traditional solar natu-ral ventilation wall,so as to provide reference and basis for improving the performance of solar ventilation wall and the indoor environment.
