Архітектура інтелектуальної системи управління ризиками та розпізнавання видів грибів
The article presents the development of an intelligent system for recognising mushroom species that provides high accuracy and ease of use. To train the model, a large dataset ‘Mushrooms classification’ from the Kaggle platform was used, which provided the necessary diversity of images and achieved...
Збережено в:
| Дата: | 2024 |
|---|---|
| Автори: | , , , , |
| Формат: | Стаття |
| Мова: | Ukrainian |
| Опубліковано: |
Vinnytsia National Technical University
2024
|
| Теми: | |
| Онлайн доступ: | https://oeipt.vntu.edu.ua/index.php/oeipt/article/view/736 |
| Теги: |
Додати тег
Немає тегів, Будьте першим, хто поставить тег для цього запису!
|
| Назва журналу: | Optoelectronic Information-Power Technologies |
Репозитарії
Optoelectronic Information-Power Technologies| Резюме: | The article presents the development of an intelligent system for recognising mushroom species that provides high accuracy and ease of use. To train the model, a large dataset ‘Mushrooms classification’ from the Kaggle platform was used, which provided the necessary diversity of images and achieved a classification accuracy of 85%. Data pre-processing included image quality checks, standardisation, and division into training, validation, and test samples, which contributed to efficient model training. The recognition algorithm is based on the ResNet convolutional neural network, which has demonstrated an accuracy advantage over other architectures. |
|---|