Thumbnail Image

Thesis

Browse

2011 - 20172
Reset filters
Subject: Others × Start date: 1990 × Type: Thesis × Thesis Type: M.Tech. ×
Reset filters

Search Results

Now showing 1 - 2 of 2
  • Thumbnail Image
    ThesisItemOpen Access
    Software development for determining quality and maturity levels of tomato using image processing techniques
    (Punjab Agricultural University, Ludhiana, 2017) Kamalpreet Kaur; Gupta, O.P.
    Agriculture contributes a lot in the economic development of India. Maturity checking has become mandatory for the food industries as well as for the farmers so as to ensure that the fruits and vegetables are not diseased and are ripe. However, manual inspection leads to human error, unripe fruits and vegetables may decrease the production. Thus, this study proposes a Tomato Classification system for determining maturity stages of tomato (Green, Breaker, Pink, Light-Red and Red- Mature) through Machine Learning using Image Processing approach. Design and development of software has been implemented using Pycharm as an IDE and Python as a programming language. The method consists of image collection, preparing database and training seven different classifiers on 80% of the total data for evaluating the maturity stages of tomato using the surface color as an attribute. Rest 20% of the total data is used for the testing purpose. The results are obtained in the form of Learning Curve, Confusion Matrix and Accuracy Score. The Graphical User Interface (GUI) for Tomato Classification by using Python is achieved. It is observed that out of seven classifiers, Random Forest is successful with 92.49% accuracy in evaluating the maturity category of tomato. It is concluded from the results that the performance of the classifier depends on the size and kind of features extracted from the data set.
  • Thumbnail Image
    ThesisItemOpen Access
    DRYING KINETICS OF GINGER (Zingiber officinale)
    (PAU Ludhiana, 2011) Sumit Maini; Ashok Kumar
    The experiments were carried out to study the drying behaviour of fresh ginger rhizome of Kozhikallan variety. The surface material was first peeled off. The material was then given various chemical treatments like hot water blanching, dipping in citric acid solution, dipping in lime and treating the product with KMS solution. The treated product was dried at 500C, 600C, 700C in the electrical tray dryer. Quality parameters of the dried product i.e. colour, texture, rehydration ratio and volatile oil content were found out. Dehydration characteristics revealed that drying took place in the falling rate period. It was also found out that pre treatment did not effect the quality much but the temperature did. Drying at 500C did not affect the quality significantly, but as the temperature increased the quality of the product significantly reduced. Volatile oil was recorded to be the highest i.e.2.2% for the product pre treated with citric acid and dried at 500C. The odour of the product pre treated with citric acid and dried at 500C was also found to be better. Change in colour values for fresh and rehydrated product was found to be less i.e. 5.93 in case of the product pre treated with citric acid and dried at 500C. Three models i.e. Newton, page and modified page models were fitted to the drying data. According to the statistical criteria (R2, chi square & RMSE), the modified page model was found to be the best fit model to describe the drying behavior of the ginger rhizome.