|
Mobile Application for Breeding Bird Classification using Deep Learning Technique |
|---|---|
| รหัสดีโอไอ | |
| Creator | Choopan Rattanapoka |
| Title | Mobile Application for Breeding Bird Classification using Deep Learning Technique |
| Contributor | Phummiphak Promrangka, Saharat Wanthong |
| Publisher | The Association of Council of IT Deans (CITT) |
| Publication Year | 2565 |
| Journal Title | Journal of Information Science and Technology |
| Journal Vol. | 12 |
| Journal No. | 1 |
| Page no. | 37-46 |
| Keyword | Mobile application, Flutter, EfficientDet model, TensorFlow Lite, TFLite model maker library |
| URL Website | https://tci-thaijo.org/index.php/JIST |
| Website title | Journal of Information Science and Technology |
| ISSN | 2651-1053 |
| Abstract | Currently, several bird parks have open aviaries where visitors can see various bird species up close. However, visitors are occasionally unable to identify the bird species they are watching. Because the bird species found in open aviaries are not typically seen in everyday life, and there are no zoo signs inside the open aviaries. Therefore, this article proposes the design and development of a mobile application using the Flutter framework. The application can detect and classify 10 bird species using a deep learning model named EfficientDet Lite, which is created by the TFLite model maker library. The users can use a smartphone camera to examine the birds. Then, when birds are found, the application will provide users the bird information. From the experiments, we found that the EfficientDet Lite 0 gave the most suitable results for the application. The model took 62.3 ms for the inference time with the precision, recall, accuracy and F1-score of 0.94, 0.94, 0.99, and 0.94, respectively. |
