جستجوی مقالات مرتبط با کلیدواژه « میزان مواد جامد محلول » در نشریات گروه « مکانیزاسیون کشاورزی »

In this article a machine vision system and an artificial neural network (ANN) for classifying the strawberry based on maturity and shape features were used. First an image processing algorithm for extracting the color and shape features was investigated and then for grading the strawberry into three classes based on shape features and three classes of maturity based on colors features were done. The sensitivity analysis indicated that shape grading had highest sensitive to area, parameter, large and minor diameters features. Also a* and S color features had better correlation coefficient than other color features with total solid soluble and therefore were selected as supreme features for grading the strawberry based on maturity. Finally, results demonstrated that the ANN was able to classify with 94.04 and 95.14 total accuracy rate for shape and maturity grading.

The potential of Fourier Transform Near-infrared (FT-NIR) spectroscopy technique was applied to determine moisture and soluble solids content of Shahani date fruit. Spectral acquisitions were carried out in reflectance mode using a FT-NIR spectrophotometer in the range of 12000- 4000cm-1. Three different preprocessing methods namely: Multiplicative Scatter Correction (MSC), first and second derivatives were employed with their effects on the ability of predictive models compared. The most suitable predictive models resulted from Partial Least Squares (PLS) method showed coefficient of determination (R2) values of 0.977 and 0.973 and root mean square errors of prediction of 1.4% and 1.6 °Brix for the moisture and the soluble solids content, respectively. The relative high Residual Predictive Deviation (RPD) values (7.4 and 6.4 for the moisture and the soluble solids content respectively) showed FT-NIR spectroscopy technique can be applied as a nondestructive, rapid and accurate method for determination of maturity of the Shahani variety date.