Optimized Bio-Inspired Thermal Image Analysis for Mastitis Detection Using YOLOv8 Probabilistic Spiking Networks
Bibliographic record
Abstract
One of the most common and financially detrimental conditions affecting dairy cows is mastitis, which is often identified by visual inspection or somatic cell count (SCC) testing, both of which are labor-intensive, subjective, and time-consuming. Existing deep learning (DL)-based approaches, including CLE-UNet, DCYOLO, FS-YOLOv4, and YOLOv7-SVM, have improved detection accuracy but still face challenges in capturing temporal dynamics, handling noisy thermal data, and optimizing model parameters effectively. To address these limitations, this work proposes a novel YOLOv8 integrated with Probabilistic Spiking Network and Secretary Bird Optimization (YOLOv8PSN-SBO) framework for automated mastitis detection using thermal imaging. The system begins with thermal video acquisition and manual annotation of key anatomical regions, followed by adaptive iterative guided filtering (AIGF) to enhance image clarity. Multi-scale feature extraction is achieved using a Cascading Residual Graph Convolutional Network (CRGCN), while YOLOv8 performs precise spatial localization of cow eyes, udders, and head postures. These spatial outputs are then temporally processed using a Probabilistic Spiking Neural Network (PSNN) that leverages biological spiking behavior to infer thermal anomalies. The Secretary Bird Optimization Algorithm (SBOA) is employed to fine-tune parameters such as confidence thresholds, anchor box sizes, loss weights, and spiking neuron dynamics for optimal performance. The proposed YOLOv8PSN-SBO model, implemented in PyTorch and validated on 1200 labeled thermal images, significantly outperforms baseline methods, achieving a detection accuracy of 91.75%, recall of 92.31%, F1-score of 91.55%, and an mAP@0.5 of 94.60%, with a real-time processing speed of 65 FPS, establishing it as a robust and efficient solution for automated mastitis detection in dairy farming.
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 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 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 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
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".