Classification of Fruits Based on CNN, SVM and PCA

Qianming Huang

2023

Abstract

In today's time, fruits are a daily necessity for human beings and people use a lot of fruits daily. To meet the demand for fruits, the total global production of fruits in 2019 was about 740 million tonnes, according to the Food and Agriculture Organisation of the United Nations (FAO). The timely handling of these fruits is undoubtedly an important issue, especially since fruits are characterized by a short shelf life. Therefore, the use of various types of machines to process fruits has become a research direction in today's world, and this includes the recognition and classification of fruit images by machines. This paper is based on a machine learning approach to construct models from fruit image datasets. Two models are used in this paper which are the SVM model with PCA and, the CNN model. Both models obtained good classification accuracy respectively, 90% for the SVM model and 97% for the CNN model. But the SVM model training time took only 2.73s whereas the CNN model training took 120.09s. Therefore, to pursue a certain level of efficiency, SVM+PCA was chosen as the model for fruit classification in a good situation where the lights are bright and the fruits are not covered.

Paper Citation

in Harvard Style

Huang Q. (2023). Classification of Fruits Based on CNN, SVM and PCA. In Proceedings of the 1st International Conference on Data Analysis and Machine Learning - Volume 1: DAML; ISBN 978-989-758-705-4, SciTePress, pages 457-465. DOI: 10.5220/0012815000003885

in Bibtex Style

@conference{daml23,
author={Qianming Huang},
title={Classification of Fruits Based on CNN, SVM and PCA},
booktitle={Proceedings of the 1st International Conference on Data Analysis and Machine Learning - Volume 1: DAML},
year={2023},
pages={457-465},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0012815000003885},
isbn={978-989-758-705-4},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 1st International Conference on Data Analysis and Machine Learning - Volume 1: DAML
TI - Classification of Fruits Based on CNN, SVM and PCA
SN - 978-989-758-705-4
AU - Huang Q.
PY - 2023
SP - 457
EP - 465
DO - 10.5220/0012815000003885
PB - SciTePress