OpenTextile-NIR: Near-infrared hyperspectral imaging and photography dataset for optical identification of textiles

This dataset contains 71 photographic images (RGB) and near-infrared (NIR) hyperspectral imaging (HSI) data of textile samples, as well as mean RGB and NIR values of a region of interest for each sample. These images were gathered from cut garment samples, each of which was approximately 10 cm by 10 cm in size. The garments originated from post-industrial sources, mainly constituting used workwear. The composition of most of the garments were known, although the dataset also includes 3 samples with unknown composition, and 11 outlier samples whose composition is believed to be different than reported. Description of this dataset in a scientific article format has been published - please refer to it when using this dataset: Sormunen, T., Mahlamäki, E., Mäkelä, S.-M., & Mäkelä, M. (2026). Introducing OpenTextile-NIR: Near-infrared hyperspectral imaging and photography dataset for optical identification of textiles. Data in Brief. https://doi.org/10.1016/j.dib.2026.112559 Keywords: hyperspectral imaging, near infrared spectroscopy, NIR, Specim SWIR, RGB camera, textiles, garments, clothing, classification, prediction, regression, machine learning