Non-destructive apple bruise detection with Raman spectroscopy and its virtual instrumentation

A prototype automated inspection system was developed to classify apples based on bruising in real time. A Nicolet FT-Raman Spectroscope was employed to obtain apple spectra. The spectroscope utilized OMNIC E. S. P. 5.1 software. The unbruised and bruised spectra were analyzed and classified by WinDAS using canonical variate analysis (CVA) and principal component analysis (PCA) models, on both the training and testing sets. The PCA and CVA model analysis satisfactorily classified the apples by bruise, then the UNEQ class modeling was used and the square of Mahalanobis distance passed χ² test. Besides, LabWIEW was applied to develop apple grading control system, and prototype of virtual instrumental system was fabricated. The experiments show that the Raman spectroscope permits non-destructive bruise determination with good results and the virtual instrument grading system has a good accuracy.