|
Thai Herb Identification with Medicinal Properties Using Convolutional Neural Network |
|---|---|
| รหัสดีโอไอ | |
| Creator | Lawankorn Mookdarsanit |
| Title | Thai Herb Identification with Medicinal Properties Using Convolutional Neural Network |
| Contributor | Pakpoom Mookdarsanit |
| Publisher | Faculty of Science and Technology, Suan Sunandha Rajabhat University |
| Publication Year | 2562 |
| Journal Title | Suan Sunandha Science and Technology Journal |
| Journal Vol. | 6 |
| Journal No. | 2 |
| Page no. | 33 to 39 |
| Keyword | Herb Identification, Leaf Recognition, Convolutional Neural Network, VGGNet, Fast R-CNN |
| URL Website | www.ssstj.sci.ssru.ac.th |
| Website title | Suan Sunandha Science and Technology Journal (SSSTJ) |
| ISSN | 2351-0889 |
| Abstract | This paper built an intelligent computer model to identify Thai herb from a single image using convolutional neural network. Thailand is one of the world herbal sources. We used 2,700 herbal images with their medicinal properties to train the computer model that covered 11 well-known Thai herbs: Siamese Rough-bush, Cumin, Holy Basil, Sweet Basil, Cha Muang, Kaffir-lime Leaf, Siamese Morning-glory, Pandanus Leaf, Mint, Chinese Kale and Chaplu, respectively. The feature extraction framework and model architecture were done by Fast Region Convolution Neural Network (Fast R-CNN) and Visual Geometry Group Network (VGGNet) that produced the recall as higher than 0.75 and the precision as higher than 0.80. |
