Prediction of Six Products from the <i>Cucurbitaceae</i> Family Using Visible/Near-Infrared Spectroscopic Data

AYDEMİR Ö.

JOURNAL OF TESTING AND EVALUATION, cilt.51, 2022 (SCI-Expanded) identifier identifier

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 51
  • Basım Tarihi: 2022
  • Doi Numarası: 10.1520/jte20220223
  • Dergi Adı: JOURNAL OF TESTING AND EVALUATION
  • Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus, Applied Science & Technology Source, Computer & Applied Sciences, INSPEC
  • Karadeniz Teknik Üniversitesi Adresli: Evet

Özet

Recently, visible/near-infrared (Vis/NIR) spectroscopy has been used in the agricultural field, especially in the food industry, for monitoring food quality, postharvest handling of products, and identification of contamination on animal feeds, as well as prediction of a variety of fruits or vegetables. In this study, six products of the cucurbitaceous commodity, including zucchini, bitter gourd, ridge gourd, melon, chayote, and cucumber, were classified using Vis/NIR spectral data. After testing spectral data as feature, we also extracted statistical features and tested them with k-nearest neighbor, Bayes, decision tree, and support vector machines classifiers. We obtained a classification accuracy rate of 99 % on the test data by applying standard normal variate technique as a preprocessing stage. The results showed that cucurbitaceous commodity could be successfully classified using Vis/NIR spectra data.