The potency of Vis/NIR spectroscopy for classification of soybean based of colour

Abstract Soybean in various colour is easy to identify using human eyes. However, it hard perform manual method for on-line production. Therefore, detection of sorting the soybean important especially industries which require a rapid and real-time task. This research was conducted study potency modular type VIS/NIR spectroscopy at wavelength 350-1000 nm classify black, green, yellow seed flour. Principal component analysis (PCA) PCA Linear discriminant (PCA-LDA) were used based on spectra pre-processing techniques. Results showed that PCA-LDA model able seeds 97% accuracy flour 100% accuracy.