In this investigation, we present an innovative approach for the identification and quantification of transparent glass films through hyperspectral imaging (HSI). The primary objective of this research entails the development of a conversion algorithm for rendering spectral information from captured imagery, specifically within the visible light and near-infrared (NIR) regions. When applied to industrial camera-generated images, this algorithm facilitates the acquisition of pertinent spectral data. The subsequent phase of this inquiry involves the application of principal component analysis to the acquired HSI images that stem from distinct processed glass samples. This analytical process normalizes the intensity of light wavelengths that are inherent in the HSI images. We derive the simulated spectral profiles by applying the Beer–Lambert law in conjunction with the generalized inverse matrix method to the normalized HSI images. These profiles are subsequently aligned with spectroscopic data collected through microscopic imaging, culminating in the visualization of characteristic dispersion patterns. The thickness of the glass processing film is successfully rendered in a visually discernible manner by employing innovative image coloring techniques. In accordance with the empirical findings, variations in the thickness of the glass coating within the NIR-HSI domain engender notable alterations in infrared transmittance across distinct wavelengths that encompass the NIR spectrum. This phenomenon provides the basis for film thickness analysis. Remarkably, the average root-mean-square error within the NIR region only amounts to 0.02, underscoring the precision of our approach. Prospective avenues of inquiry that stem from this research include the integration of the developed methodology into the conception of a real-time, large-area automated optical inspection system.