Related Experiment Video
Updated: Aug 14, 2026
09:47
Spotting Cheetahs: Identifying Individuals by Their Footprints
Published on: May 1, 2016
Deep-learning-based buffalo identification through muzzle pattern images
Orhan Ermetin1, Humar Kahramanlı Örnek2
1Faculty of Agriculture, Department of Animal Science, Yozgat Bozok University, Yozgat, Türkiye.
Archives Animal Breeding
|August 13, 2026
Summary
Artificial intelligence (AI) enhances animal biometrics for accurate buffalo identification. SqueezeNet achieved 99.88% accuracy, demonstrating AI
Area of Science:
- Artificial Intelligence
- Animal Biometrics
- Livestock Management
Background:
- Accurate identification of buffaloes is vital for livestock producers and researchers.
- Current methods may lack efficiency for large-scale record-keeping and tracking.
- Advancements in AI offer new possibilities for animal recognition.
Purpose of the Study:
- To develop and evaluate an AI-supported system for buffalo recognition.
- To assess the performance of different AI algorithms for buffalo identification.
- To establish an effective method for identifying large livestock using facial biometrics.
Main Methods:
- Utilized facial images of 11 buffaloes from Yozgat province to create a dataset.
- Applied and compared four different artificial intelligence algorithms for recognition.
- Evaluated algorithm performance using metrics such as accuracy, precision, recall, and F1 score.
Main Results:
- All four AI algorithms demonstrated successful buffalo recognition.
- SqueezeNet achieved the highest performance with 99.88% accuracy, 0.998 precision, 0.999 recall, and 0.999 F1 score.
- ResNet101 showed the lowest performance among the tested algorithms, with 99.30% accuracy.
Conclusions:
- AI-based facial recognition is a viable and highly accurate method for buffalo identification.
- SqueezeNet and GoogLeNet algorithms show exceptional performance for this application.
- The developed system can significantly aid in livestock management and research record-keeping.