Information About the Undergraduate Thesis Project
This web application was developed as part of an undergraduate thesis entitled "Classification of Coccinellidae Beetle Species Based on Photographic Images Using the EfficientNetV2 Architecture." The research aims to classify ten species of Coccinellidae beetles automatically from photographic images using a deep learning approach based on the EfficientNetV2 architecture. The system is designed to assist species identification by providing fast, accurate, and user-friendly classification through a web-based interface.
| Research Title | Classification of Coccinellidae Beetle Species Based on Photographic Images Using the EfficientNetV2 Architecture |
|---|---|
| Researcher | Riolan Pratama |
| Supervisors |
Ernawati, S.T., M.Cs. Prof. Agustin Zarkani, S.P., M.Si., Ph.D. |
| Study Program | Informatics Engineering |
| Deep Learning Architecture | EfficientNetV2 |
| Research Year | 2026 |
| Number of Species | 10 |
|---|---|
| Images per Species | 250 |
| Total Images | 2,500 |
| Image Resolution | 512 × 512 pixels |
| Dataset Source | iNaturalist (photographic images) with additional hand-drawn sketch images |
| Image Type | Photographic Images and Hand-Drawn Sketches |
| Deep Learning Architecture | EfficientNetV2 |
|---|---|
| Framework | TensorFlow / Keras |
| Transfer Learning | Yes |
| Fine-Tuning | Yes |
| Input Image Size | 512 × 512 pixels |
| Prediction Output | Species Name and Confidence Score |
| Python | Main programming language |
|---|---|
| Flask | Web application framework |
| HTML5 | Structure of the web pages |
| CSS3 | Styling and responsive layout |
| Bootstrap 5 | Responsive user interface |
| JavaScript | Interactive client-side functionality |
| TensorFlow / Keras | Deep learning framework |
| EfficientNetV2 | CNN architecture for image classification |
| NumPy | Numerical computation |
| Pillow (PIL) | Image processing |
This research was conducted as part of the undergraduate thesis in the Informatics Engineering Study Program. The author sincerely expresses gratitude to the supervisors for their invaluable guidance, support, and encouragement throughout the research process. Appreciation is also extended to everyone who contributed directly or indirectly to the completion of this research and the development of this web-based classification system.