Freshness and quality assessment of parsley using image processing and artificial intelligence techniques

Authors

  • Mohammad Hosseinpour Zarnaq Department of Agricultural Machinery Engineering, University of Tehran, Iran
  • Mahmoud Omid Professor, Department of Agricultural Machinery Engineering, Faculty of Agricultural Engineering and Technology, University of Tehran, Karaj, Iran
  • Mahmoud Soltani Firouz Assistant Professor, Department of Agricultural Machinery Engineering, Faculty of Agricultural Engineering and Technology, University of Tehran, Karaj, Iran
  • Mostafa Jafarian Ph.D. graduate, Department of Mechanical Engineering of Agricultural Machinery, Samangan Faculty of Agriculture, Technical and Vocational University (TVU), North-Khorasan, Iran.
  • Pourya Bazyar MSc graduated, Department of Agricultural Machinery Engineering, Faculty of Agricultural Engineering and Technology, University of Tehran, Karaj, Iran.

Abstract

Fruits and vegetables are important components of healthy diets. Vegetable freshness is important for both postharvest industry and consumer appeal. This study focused on freshness detection of parsleys using combined image processing and artificial intelligence techniques. A dataset of color and texture features computed from parsley images. Linear discriminant analysis (LDA) and principal component analysis (PCA) methods are used for feature reduction. Multilayer perceptron (MLP) neural networks, support vector machine (SVM) and decision trees (DTs) classifiers were used for classification. Results showed MLP with LDA feature selection methods had higher performance and the overall accuracy, RMSE, MAE of MLP classifier (using LDA feature selection) were 97.22%, 0.17, and 0.03, respectively. This approach provided a rapid and nondestructive detection of parsley freshness without using chemical or colorimetric analysis. The results demonstrated that suggested approach could be employed satisfactorily for inspection, classification and automation of vegetables postharvest operations.

Downloads

Published

2022-06-28

Issue

Section

VI-Postharvest Technology and Process Engineering