In a typical forensic investigation, fabric analysis plays a vital role in solving different crimes. Several types of textile fibre materials (cotton, polyester, denim, polypropylene, polycotton, and viscose) were analysed in the presence of common contaminants (blood, rainwater, seawater, sand and gunshot-residue) to evaluate the performance of two NIR spectrometers for in situ analysis of different crime scene conditions. The spectrometers used were SCIO® by Consumer Physics and NIRscan Nano by Texas Instruments. The SCIO instrument covers the third overtone region (740–1070 nm) and NIRscan Nano instrument covers the first and second overtone regions (900–1700 nm). Spectra from both instruments were pre-processed using the PRFFECTv2 software to eliminate noise and smooth the data for classification model construction. The models showed high accuracy, sensitivity and specificity with a range of 69–100% for binary classification (one class versus others) and range of 76–100% for multi-class classification of fibre material. This study shows for the first time the capability of pocket-size spectrometers coupled with random forest models for classification of fibre material in the presence of common contaminants in a rapid and non-destructive manner, and so is suitable for investigation of crime scenes.