A sample of front profile images of 50 cattle, with 50 images per cattle, facilitating research in cattle facial recognition, breed classification, and machine learning algorithms for cattle facial feature analysis
The Cattle Face Images Dataset is a collection of recorded videos of 50 cattle. With 50 images per cattle, this dataset provides comprehensive coverage of cattle facial features from the front profile. The dataset contains two extra folders from the original videos, where the face pictures were detected and extracted. Each cattle are assigned a unique number, and there are 50 folders, one for each, making it easy to navigate and locate specific cattle images. Researchers in computer vision, machine learning, and animal sciences can utilize this dataset for various applications, including facial recognition systems, breed classification, biometric identification, and developing machine learning algorithms for cattle facial feature analysis. The images in the dataset offer a clear and detailed view of the frontal aspect of each cattle's face, enabling the study of unique characteristics and variations in facial structures. The cattle breeds represented in the dataset include Holstein, Hereford, and Jersey. Accompanying labels and metadata provide essential information such as cattle identifier, breed, and any additional relevant details. These facilitate consistency and ease of use for researchers exploring cattle facial features and related applications. We provide the Cattle Face Images Dataset on GitHub to encourage collaboration and facilitate research in machine vision, pattern recognition, and animal sciences. It aims to support the development of accurate systems for cattle identification, monitoring, and management. Please note that the Cattle Face Images Dataset is intended for research and educational purposes, and it is made available under an open license to foster knowledge sharing within the scientific community.
A sample of lablled dataset for the detection model also is provided.
Please Cite:
@article{bakhshayeshi2023intelligence, title={An Intelligence Cattle Re-Identification System over Transport by Siamese Neural Networks and YOLO}, author={Bakhshayeshi, Ivan and Erfani, Eila and Taghikhah, Firouzeh Rosa and Elbourn, Stephan and Beheshti, Amin and Asadnia, Mohsen}, journal={IEEE Internet of Things Journal}, year={2023}, publisher={IEEE} }
Folder Structure: Cattle_1 Cattle_2 ... Cattle_50 Extra_Folder_1 Extra_Folder_2