Submitted to: IFAC/CIGR Workshop
Publication Type: Abstract Only
Publication Acceptance Date: 1/18/2006
Publication Date: 3/26/2006
Citation: Kavdir, I., Lu, R., Ariana, D., Ngouajio, M. 2006. Nondestructive quality assessment of pickling cucumbers using visible and near infrared spectroscopy [abstract]. IFAC/CIGR Workshop. Paper No. CAPPT0068. Interpretive Summary:
Technical Abstract: Firmness, skin and flesh color, and dry matter content are important quality parameters for pickling cucumbers. Nondestructive measurement of these quality parameters could provide a means for sorting and grading pickling cucumbers to ensure the quality and consistency of final pickled products. This study was aimed at developing a nondestructive method for measuring the firmness, skin and flesh color, and dry matter content of pickling cucumbers by means of visible and near-infrared (Vis/NIR) spectroscopy. 'Journey' and 'Vlaspik' pickling cucumbers were hand harvested, respectively, from an experimental field at Michigan Sate University Horticulture Teaching and Research Center (East Lansing, MI, USA) and a commercial field in Ravenna, Michigan in 2005. After harvest, the cucumbers were stored in two different conditions (at 20 deg C and 95% relative humidity and at 10 deg C and 95% relative humidity) for two weeks. Seventy-five cucumbers from each storage condition were tested on alternate days during the two-week storage period. Approximately 2000 cucumbers from the two varieties were tested in the study. Nondestructive spectral measurements and destructive physical properties measurements were taken from the same location (i.e., the middle section) of the cucumbers. Spectroscopic measurements were made from each cucumber in interactance mode with a miniaturized CCD-based Vis/NIR spectrometer (Model S2000, Ocean Optics Inc., Florida, USA) which had a spectral range between 550 nm and 1100 nm. Skin and flesh color was measured from the intact and peeled cucumbers using a digital colorimeter. Fruit firmness was tested using a texture analyzer with a 6-mm diameter probe to penetrate the peeled fruit 10 mm. The maximum force and strain energy (i.e., the area under the force/deformation curve) recorded were considered to be measures of fruit firmness. The dry matter content of the cucumbers was determined by drying cucumber slices in a convection oven at 80 deg C for 24 h. Relative reflectance spectra were calculated from the raw reflectance spectra of cucumber samples and preprocessed using methods such as Savitzky-Golay smoothing, first derivative and mean-centering. The partial least squares method was used to develop calibration models from 2/3 of the cucumber samples for predicting firmness, skin and flesh chroma, and dry matter content. The models were validated with the remaining samples. Good correlations were obtained in predicting firmness values with the coefficient of determination (R**2) between 0.70-0.75 for calibration and between 0.62-0.65 for validation. Better correlations were obtained in predicting color values (R**2 between 0.85-0.89 for calibration and between 0.75-0.78 for validation). Due to a narrow range of the dry matter content among the cucumber samples (with 4.0% mean and 0.41% standard deviation), poor predictions were obtained with Vis/NIR spectroscopy. The best results reported here were obtained taking natural logarithm of the spectra and using all the cucumber data from both sources (varieties) and stored at 10 deg C. In addition to the logarithm of the spectra, first and second derivatives were also taken for the purpose of finding optimal results. This research demonstrated the potential usefulness of Vis/NIR spectroscopy for measuring quality of pickling cucumbers. However, further studies are needed on different light illumination/sensing modes in order to improve the performance of Vis/NIR spectroscopy in predicting quality of pickling cucumbers.