Cattle AutoID: Biometric for Cattle Identification.

Kusakunniran W, Phongluelert K, Sirisangpaival C, Narayan O, Thongkanchorn K, Wiratsudakul A. Cattle AutoID: Biometric for Cattle Identification. SIET '23: Proceedings of the 8th International Conference on Sustainable Information Engineering and Technology. 2023;570-574. https://doi.org/10.1145/3626641.3627215

Summary: This study introduces a biometric method for cattle identification using facial images, addressing limitations of traditional RFID, ear tags, and microchips. The system uses convolutional neural networks (CNN) for both face localization and recognition, employing Single-Shot Detector (SSD) for face localization and FaceNet for recognition. The method was tested on a dataset of 2,432 images from 152 cattle, achieving subject-based accuracy of 94.74% and image-based accuracy of 83.45%. This approach provides a non-invasive, cost-effective alternative to current cattle identification methods.