발간년도 : [2023]
| 논문정보 |
|
| 논문명(한글) |
[Vol.18, No.5] A System for Determining Tomato Maturity Based on Deep Learning |
|
| 논문투고자 |
Jae-Heon Kim, Dae-Sung Kim, Meong-Hun Lee |
|
| 논문내용 |
Currently, farmers are showing problems such as aging and rising labor costs, and the development of smart agriculture that combines cutting-edge technologies such as AI, video analysis, and big data is essential to solve these problems. In this study, a deep learning-based tomato maturity determination system was studied for the development of smart agriculture. In order to preprocess tomato images, colors were extracted for each image with RGB and HSV color models, and the extracted images were preprocessed into images that were easier to learn through noise removal using Gaussian filters and data normalization through Standard Scaler. The pretreated tomato image learned a tomato maturity discrimination image according to color using a deep learning model called ResNet-50 and a tomato maturity discrimination model was obtained. If a tomato maturity determination model is stored separately and then combined with a camera to take an image, the rating of the captured tomato can be immediately checked, and the captured image is also stored in the database for future model updates. It is expected to contribute to the great development of smart agriculture by using this tomato maturity determination system to prevent damage that may occur due to missed harvest time and to produce a system that can determine maturity by using the system. |
|
| 첨부논문 |
|
|
|
|
|
자료실 > 논문지 
