Earticle

초록

영어

In this study, we analyzed animal registration data to identify the most popular dog breeds raised in South Korea. And then, a dataset was collected for the identified dog breeds and used to perform transfer learning on the YOLOv8 model to develop a breed classification model, and the classification accuracy was measured for each dog breed. The accuracy of classifying dog breeds by breed was confirmed to be at least 84% and up to 100%.

목차

Abstract
I. INTRODUCTION
II. RELATED WORKS
A. Dog Breed Classification Dataset
B. Related works
C. YOLOv8
III. METHOD
A. Data Preparation
B. Collect data
C. Refine and process data
IV. TRAINING MODEL
A. YOLO model and hyperparameter
B. Results
V. IMPLEMENTATION
VI. CONCLUSION
A. Analysis result
B. Future work
ACKNOWLEDGMENT
REFERENCES

저자

• Jeonghyun Choi [ AI Convergence Research Center Sejong University ]
• Daeyong Kim [ The Little Cat The Little Cat Seoul, Republic of Korea ]
• Chinchol Kim [ AI Convergence Research Center Sejong University ]
• Seungwoo Choi [ AI Convergence Research Center Sejong University ]
• Jaeyoo Lee [ AI Convergence Research Center Sejong University ]
• Seongjoon Yoo [ AI Convergence Research Center Sejong University ] Corresponding Author

참고문헌