Apple Classification Based on HOG, KNN and SVM
Jiahui Huang
2023
Abstract
In the rapid development of deep learning, traditional machine learning in the field of classification has the advantages of simplicity, ease of understanding, and strong interpretability. Apples, as an important global agricultural product, bring a lot of economic value and have health benefits for human beings. However, they are time-consuming and labour-intensive to sort manually. Therefore, realizing the intelligence of classification process is helpful to improve economic efficiency. For the apple dataset with high similarity, this research adopts two models, k-nearest neighbor (KNN) and support vector machine (SVM), and combines four models with two features, Histogram of Oriented Gradients (HOG) feature extraction and original features, to compare and research the models suitable for apple classification. It is found that HOG features do not perform well on apple images of similar shape and size, but both SVM and KNN using raw features show good performance on both training and test sets. The proposed method is simple to implement, has high accuracy and is suitable for further extension of application to other fruit domains.
DownloadPaper Citation
in Harvard Style
Huang J. (2023). Apple Classification Based on HOG, KNN and SVM. In Proceedings of the 1st International Conference on Data Analysis and Machine Learning - Volume 1: DAML; ISBN 978-989-758-705-4, SciTePress, pages 451-456. DOI: 10.5220/0012814800003885
in Bibtex Style
@conference{daml23,
author={Jiahui Huang},
title={Apple Classification Based on HOG, KNN and SVM},
booktitle={Proceedings of the 1st International Conference on Data Analysis and Machine Learning - Volume 1: DAML},
year={2023},
pages={451-456},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0012814800003885},
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 - Apple Classification Based on HOG, KNN and SVM
SN - 978-989-758-705-4
AU - Huang J.
PY - 2023
SP - 451
EP - 456
DO - 10.5220/0012814800003885
PB - SciTePress