Exploring Key Visual Features for Early Lameness Detection: Toward Transparent Intelligence

Lameness in cattle, characterized by abnormal stride and gait, poses significant economic and welfare challenges in agriculture. Traditional visual inspections lack accuracy and scalability, prompting the development of transparent computer vision-based detection systems. This study leverages a dataset of ۱۷۰ cattle videos from public sources and the University of Tehran’s Cattle Farm, preprocessed into ۱۲۲۶ one-second sub-clips (۴۱۶×۴۱۶ pixels, ۲۵ FPS) to mitigate noise from unpredictable cattle behavior. Using the YOLOv۷ model, we extracted ۳۵ temporal features, including step sizes, speed, acceleration, and relative head-to-leg coordinates, focusing on the cattle’s head, legs, and back. These features were further engineered using time-series characterization techniques and hypothesis testing, yielding ۳۷۷۳ features. A deep learning model, trained on these features, achieved ۸۸.۶۶% accuracy and ۹۳.۷۴% AUC, while a Light Gradient Boosting Machine model on engineered features reached ۸۱.۳% accuracy and ۹۰.۸% AUC. Sensitivity analysis highlighted leg and head-related features as critical for lameness detection. By emphasizing interpretable features and robust modeling, this approach enhances transparency, improving animal welfare and farm productivity under diverse conditions.

کلیدواژه ها:

Multivariate Time-Series Analysis ، Feature Engineering ، Deep Learning ، Machine Learning ، Animal Health and Welfare