Enhanced you only look once model with frequency feature enhancement and illumination perception for duck behavior recognition in dynamic light scenarios
Bibliographic record
Abstract
Modern intensive duck farming has improved production efficiency while facing health problems for ducks. As duck behaviors are closely related to their health status, accurately monitoring their behaviors is necessary. With the development of artificial intelligence (AI), the application of AI offers an effective approach to animal behavior recognition. Currently, accurately recognizing duck behaviors consistently from day to night remains a challenge. This challenge stems from the persistent dynamic changes in light, which can lead to significant performance degradation in conventional behavior recognition methods. To overcome this challenge, this study proposes an enhanced You Only Look Once version 11 small (YOLOv11s) with frequency feature enhancement and illumination perception for duck behavior recognition in dynamic light scenarios (FIY4DBR). Specifically, to tackle the problem of unclear edge, texture, and behavior characteristics of ducks in the low-light condition, a frequency feature enhancement mechanism (FFEM) is designed, which effectively enhances the duck feature representation ability. Additionally, to improve the model’s robustness to light variations, an illumination perception mechanism (IPM) is developed, which adjusts the contrast of objects and background features according to different brightness conditions, thereby enhancing the model’s generalization capability across different brightness scenarios. Experimental simulations on the self-built dataset show that FIY4DBR achieves an average recognition precision of 92.0% and recall of 88.8%, representing improvements of 1.8 and 3.6 percentage points over the baseline YOLOv11s. This demonstrates that the proposed FIY4DBR provides a high-precision and highly adaptive solution for intelligent livestock behavior monitoring, contributing to advancing the development of intelligent farming technologies.
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 distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| 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 | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 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 teacher head, 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".