|
Image based durian (Durie azomethines Linn) sweetness measurement by ResNet50 |
|---|---|
| รหัสดีโอไอ | |
| Creator | Chomtip Pornpanomchai |
| Title | Image based durian (Durie azomethines Linn) sweetness measurement by ResNet50 |
| Publisher | Faculty of Engineering, Khon Kaen University |
| Publication Year | 2567 |
| Journal Title | Engineering and Applied Science Research |
| Journal Vol. | 51 |
| Journal No. | 2 |
| Page no. | 200-210 |
| Keyword | Convolutional neural network, Pattern recognition, ResNet50, Sweetness measurement |
| URL Website | https://ph01.tci-thaijo.org/index.php/easr/index |
| Website title | Engineering and Applied Science Research |
| ISSN | 2539-6161 |
| Abstract | Development of the capability to determine durian sweetness using a single image is the main objective of this research. The developed system is called the “Durian Sweetness Measurement System” or DSMS. The DSMS employed ResNet50 in the MATLAB toolbox to recognize durian imagery. The system consists of four main subprograms, 1) durian dataset creation, 2) image acquisition, 3) durian sweetness evaluation, and 4) results illustration. The system was used to conduct experiments on 17 Monthong durian pulps in 102 video clips. The DSMS determined that sweetness of the raw, mature and ripe durian was around 14–19, 20–26 and 27–31 oBrix, respectively. The accuracy of the DSMS is 97.57%, with an average access time of 1.5248 sec per image. |
