Air Pollution: Sensitive Detection of PM2.5 and PM10 Concentration Using Hyperspectral Imaging

This paper proposes a method to detect air pollution by applying hyperspectral imaging algorithm for visible light, near infrared, and far infrared. By assigning information images from monocular, thermal imaging, principal component analysis is performed on taken at different times obtain the solar radiation intensity. The Beer–Lambert law multivariate regression are used calculate PM2.5 PM10 concentrations during period, which compared with corresponding Taiwan Environmental Protection Agency evaluate accuracy of this method. study reveals that in light band higher than near-infrared far-infrared bands, it also most convenient data acquisition. Therefore, future, mobile phone cameras will be able analyze any given time using capturing increase convenience immediacy detection.