1. درجهبندي بصري گوجهفرنگی گیلاسی پوششدار شده با ژل آلوئهورا حاوي روغن شاهدانه با روشهاي تجزیه و تحلیل مولفههاي اصلی، شبکه عصبی مصنوعی و سیستم استنتاج تطبیقی عصبی-فازي.
- Author
-
علی گنجلو, محسن زندي, ماندانا بی مک, and سمانه منجم
- Subjects
- *
ARTIFICIAL neural networks , *OILSEEDS , *PRINCIPAL components analysis , *IMAGE processing , *IMAGING systems , *EDIBLE fats & oils - Abstract
In the present study, in the first step, the effect of Aloe vera gel (75% v/v) coating containing different concentrations of hemp seed oil (1-5% v/v) on some physicochemical properties of cherry tomatoes during storage at room temperature was investigated. The results revealed the ability of hemp seed oil to improve the physicochemical properties of cherry tomatoes during storage, although no significant difference was observed between 3 and 5% levels of hemp seed oil (p> 0.05). Slope change in the ripening index trend occurred for A. vera gel (75% v/v) coated sample on day 12 and for A. vera gel containing 3% hemp seed oil coated sample on day 16. Using an image processing system, the changes of the coated samples were evaluated based on the color statistical and color texture features extracted from the images and were graded through different procedures. The results showed that the principal component analysis (PCA) and artificial neural network (ANN) methods were able to divide the cherry tomatoes into intact and blemished grades which the ANN method was graded samples using color texture features with higher accuracy (97.41%). The adaptive neuro-fuzzy inference system (ANFIS) method had higher diagnostic power than the other two methods and was able to grade the samples into three grades including intact, grade 2 and unusable with accuracy of 98.96%. [ABSTRACT FROM AUTHOR]
- Published
- 2021
- Full Text
- View/download PDF