DIH4AI: Hyperspectral Datasets for Denim Fabric Composition OC2

This dataset contains hyperspectral images of denim fabric over 224 reflectance bands. It was used to test automated composition analysis in the DIH4AI Open Call 2 FABCOD project (Fabric Composition Detector).

Dataset

FABCOD Datasets

License

Creative Commons Attribution 4.0 International License

Main Characteristic

Data was acquired using a 900-1700nm hyperspectral linear camera, 224 bands. Data resources are divided into two categories, 100% cotton and cotton + elastane.

Format: ENVI binary data (.raw) representing raw reflectance images (640px, 224 bands); .hdr file containing image metadata; or .mat file. In this latter case, images are already calibrated, whereas the former include via "dark" and "light" calibration reference files.

Research areas

Physical AI

Technical Categories

Computer vision Machine learning Robotics and automation

Business Categories

Manufacturing

Keywords