基於影像和環境資料的多模態資料分析技術進行大氣能見度和PM2.5之分類效能探討

能見度是衡量觀測點大氣透明度的指標，以人們能夠看到並識別物體的最大水平距離來表示。低大氣能見度常與空氣污染伴隨發生，對交通安全和人體健康產生危害。本研究利用衛星影像與環境資料，探討兩種不同多模態資料分析技術對能見度和PM2.5分類效能的影響。第一種方法使用深度學習開發四種多模態資料分類模型。第二種方法結合深度學習和機器學習，創建了十二種多模態資料分類器。根據實驗結果，加入各種環境資料顯著提升了衛星影像的分類效能。在進行能見度和PM2.5分類任務時，使用結合深度學習和機器學習的多模態技術可以獲得最佳分類效能，測試準確率分別達到0.978和0.957。在環境資料中，天氣特徵顯著影響能見度，但污染物濃度對其影響有限。對於PM2.5，加入天氣特徵後的效能提升不如能見度分類來的明顯，需結合其他空氣品質特徵以改善分類效能。

Visibility is a measure of the atmospheric transparency at an observation point, expressed as the maximum horizontal distance over which a person can see and identify objects. Low atmospheric visibility often occurs in conjunction with air pollution, posing hazards to both traffic safety and human health. This study combined satellite images with environmental data to explore the classification performance of two distinct multimodal data analysis techniques. The first approach involves developing four multimodal data classification models using deep learning. The second approach integrates deep learning and machine learning to create twelve multimodal data classifiers. Experimental results show that the inclusion of various environmental data significantly enhances the classification performance of satellite imagery. For visibility and PM2.5 classification, utilizing multimodal techniques combining deep learning and machine learning achieves optimal classification performance, with test accuracies reaching 0.978 and 0.957, respectively. Weather features notably affect visibility within environmental data, while pollutant concentrations have a limited impact. For PM2.5, the performance improvement from adding weather features is less pronounced compared to visibility, necessitating the inclusion of air quality features to enhance classification performance.