|
Comparison the Performance of Models for Classifying Thai Exported Fruit Images |
|---|---|
| รหัสดีโอไอ | |
| Creator | Walaiporn Sornkliang |
| Title | Comparison the Performance of Models for Classifying Thai Exported Fruit Images |
| Contributor | Woraphat Jaihao, Prumrapee Thongkaew, Thanabdin Hemtanon, Pichayut Srinuankaew, Kritaphat Songsri-in |
| Publisher | Faculty of Informatics, Mahasarakham University |
| Publication Year | 2568 |
| Journal Title | Journal of Applied Informatics and Technology |
| Journal Vol. | 7 |
| Journal No. | 1 |
| Page no. | 73-90 |
| Keyword | Image Classification, Artificial Neural Network, Exported Fruits |
| URL Website | https://ph01.tci-thaijo.org/index.php/jait |
| Website title | Journal of Applied Informatics and Technology |
| ISSN | 3088-1803 |
| Abstract | Fruits are a major economic crop in Thailand, providing both primary employment and income for many farmers. This research aims to develop and evaluate the performance of a classification model for Thai export fruits. A dataset of over 710 images of Thai export fruits was collected. The model learning process consists of two steps: dimensionality reduction of image data and model learning. In the dimensionality reduction step, two methods were experimented with: image squashing and convolutional neural network. The models studied in the second step consisted of three models: the nearest neighbor model, the decision tree model, and the logistic regression model. The research results on the collected dataset found that the image classification performance of Thai export fruits using dimensionality reduction techniques with convolutional neural networks is higher than that of image flattening, and the model that can best classify images is the logistic regression model, which achieves a maximum accuracy and recall of 95.21%, a precision at 95.50%, F1 score at 95.16%. It used the most training time of 25.809 seconds but only took 7.70 milliseconds for inference. |
