Dairy DigiD ∼ A Deep Learning-Based, Non-Invasive Biometric Identification System for Dairy Cattle Using Detectron2
Bibliographic record
Abstract
Abstract Precision livestock farming demands accurate and reliable individual animal identification for optimizing health monitoring, resource allocation, and overall herd productivity. Conventional identification methods—such as ear tagging and wearable sensors—are often invasive, potentially causing animal discomfort and affecting natural behaviors. To address these challenges, we introduce Dairy DigiD, a non-invasive biometric classification system for dairy cattle using facial image analysis. Leveraging deep learning and the Detectron2 framework, Dairy DigiD classifies cattle into four categories—young (<2 years), mature milking, pregnant, and old—using over 2,500 high-resolution facial images. A DenseNet121 model achieved a classification accuracy of 97%, showcasing its strong discriminative capability. Further, Detectron2 attained an average recall of 87% in category identification, an average precision of 96% in cattle detection, and an overall detection accuracy of 93%, demonstrating its robustness and adaptability to varied environmental conditions. Unlike traditional approaches that rely on fixed facial landmarks and can be sensitive to environmental variability, Dairy DigiD’s flexible deep learning architecture enables stable and scalable facial feature extraction. This non-invasive method minimizes animal stress, enhances data integrity, and facilitates improved herd management and welfare monitoring. In comparison to existing convolutional neural network (CNN)-based strategies, Dairy DigiD exhibits superior adaptability to real-world farm conditions, including variable lighting and complex poses. This work highlights the potential of integrating advanced computer vision and deep learning methodologies into precision dairy farming, offering an ethically aligned, efficient, and scalable approach to modern livestock management.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.004 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".