1. شناسايي و تخمين ميزان فرايند دانهسازي در پرتقال رقم والنسيا با استفاده از طيفسنجي مادونقرمز نزديکو ماشين بردار پشتيبان
- Author
-
محسن بهپور بلهسور and رحمان فرخي تيمورلو
- Subjects
FEATURE extraction ,OPTICAL spectroscopy ,SUPPORT vector machines ,GRANULATION ,NEAR infrared spectroscopy - Abstract
The physiological disorder of citrus granulation is one of the qualitative issues in consumer markets that is not visually detectable. One of the applicable methods for the non-destructive study of the internal tissue of agricultural products is the visible and near-infrared spectroscopy technique. In this research, the VIS/NIR signals in the wavelength range of 200-1100 nanometers were recorded for each sample. Spectroscopy was performed on the samples With contact. Then, the amount of reflectance was calculated in the interaction mode, and absorption spectra were normalized using the Min-Max method. Subsequently, the recorded spectra were smoothed by the moving average (MA) method, the Stavisky-Golay algorithm was applied to each spectrum, and seven statistical features were extracted from each spectrum. Five levels were defined to show the intensity and amount of granulation in the internal tissue of oranges, including levels A, B, C, D, and E. The degree of granulation of each sample was defined based on the dryness and appearance of the dried area using a destructive method. Support vector regression (SVR) and support vector machine (SVM) were used to estimate oranges' moisture content and detect granulation levels in oranges, respectively. The results showed that with the increase and development of the intensity of granulation, the juice bags become harder, drier, and bigger. Their water content decreased; the highest moisture content of oranges was 90.97% in the state without granulated lesions, and the lowest was 83.36% in the state where more than 75% of its tissue was granulated. Also, the examination of ViS-NIR spectra showed that with the development of granulation in oranges, the absorption in the 400-950 nanometer range significantly decreased, and the difference in absorption intensity for different levels of granulation was maximum in the 570-850 nanometer range. Granulation level detection results showed that the overall accuracy of the support vector machine for linear, polynomial, and Gaussian radial kernels was 92.50%, 96.50%, and 95.00%, respectively. The sensitivity of SVM with polynomial kernel in detecting the levels A, B, C, D, and E granulation were equal to 98.0%, 91.40%, 97.30%, 96.80%, and 95.70%, respectively. [ABSTRACT FROM AUTHOR]
- Published
- 2024
- Full Text
- View/download PDF