Journal of Environmental Biology

Determination of olive cultivars by deep learning and ISSR markers

 ????????????????????

M. Sesli¹*, E.D. Yegenoğlu² and V. Altıntas³ ??????? 

¹Manisa Celal Bayar University, School of Tobacco Expertise, Akhisar, Manisa, 45200, Turkey

²Manisa Celal Bayar University, Alasehir Vocational School, Alasehir, Manisa, 45600, Turkey

³Manisa Celal Bayar University, Akhisar Vocational School, Akhisar, Manisa, 45200, Turkey

*Corresponding Author Email : meltem.sesli@cbu.edu.tr

Paper received: 29.04.2019 ?????? ???????????????????????????????????????Revised received: 29.09.2019 ???????????? ????????????????????????????????Accepted: 14.12.2019

Abstract

Aim: The aim of the study was to make accurate estimation of olive varieties by using morphologic characters through deep learning and genetic characters through ISSR (Inter Simple Sequence Repeats) markers.

Methodology: In this study, 800 leaf samples were collected from olive varieties and training and testing was performed; 600 samples were assessed for the training process and 200 samples were assessed for the testing process. Convolution of neural networks is a component of deep learning which is used frequently in image processing was used in this study.

Results: Based on the results of such classification, the designed model was successful at a rate of 89.57% and it was also determined that this structure can be used in the area of problem.

Interpretation: The success of convolution neural networks in terms of classification was exhibited. In ISSR method, the evaluation was performed on the basis of DNAs, i.e., genetic properties of varieties by means of ISSR markers.

Keywords: Deep learning, ISSR markers, Neural networks, Olive cultivars

Copyright ? 2020 Triveni Enterprises. All rights reserved. No part of the Journal can be reproduced in any form without prior permission. Responsibility regarding the authenticity of the data, and the acceptability of the conclusions enforced or derived, rest completely with the author(s).