Hyperspectral imaging combined with PCA for discrimination of chemical pills

Hyperspectral imaging (HSI) combines conventional imaging and spectroscopy to attain both spatial and spectral information from an object. Spectral information is allowed identification and quantification of chemical components within a sample. The proposed system takes an input, color image of chemical pills. The process in general involves segmentation, feature extraction and discrimination. In this paper, feature extracted from the segmented pills are motivated by the concept of the well-known principal component analysis (PCA). To use quantitative evaluation of the patterns as the object's features build a principal component space for all object's pixels. The hyperspectral images were obtained using Spectral dimensions MatrixNIR chemical imaging system with a spectral range 960–1662 nm, spatial resolution 320 × 256 pixels.