Jeon Ryan L, Peschel Joshua M, Ramirez Brett C, Stock Joseph D, Stalder Kenneth J
Department of Agricultural and Biosystems Engineering, Iowa State University, Ames, IA 50011, US.
Department of Animal Science, Iowa State University, Ames, IA 50011, US.
Transl Anim Sci. 2023 Mar 21;8:txad033. doi: 10.1093/tas/txad033. eCollection 2024.
This paper presents a visual deep learning approach to automatically determine hock and knee angles from sow images. Lameness is the second largest reason for culling of breeding herd females and relies on human observers to provide visual scoring for detection which can be slow, subjective, and inconsistent. A deep learning model classified and detected ten and two key body landmarks from the side and rear profile images, respectively (mean average precision = 0.94). Trigonometric-based formulae were derived to calculate hock and knee angles using the features extracted from the imagery. Automated angle measurements were compared with manual results from each image (average root mean square error [RMSE] = 4.13°), where all correlation slopes (average = 0.84) were statistically different from zero ( < 0.05); all automated measurements were in statistical agreement with manually collected measurements using the Bland-Altman procedure. This approach will be of interest to animal geneticists, scientists, and practitioners for obtaining objective angle measurements that can be factored into gilt replacement criteria to optimize sow breeding units.
本文提出了一种视觉深度学习方法，用于从母猪图像中自动确定跗关节和膝关节角度。跛行是繁殖 herd 雌性淘汰的第二大原因，依靠人类观察者进行视觉评分检测，这可能会很慢、主观且不一致。一个深度学习模型分别从侧面和后面轮廓图像中分类并检测出10个和2个关键身体地标（平均精度 = 0.94）。利用从图像中提取的特征，推导了基于三角学的公式来计算跗关节和膝关节角度。将自动角度测量结果与每张图像的手动测量结果进行比较（平均均方根误差[RMSE] = 4.13°），所有相关斜率（平均 = 0.84）在统计学上均与零有差异（< 0.05）；使用Bland-Altman程序，所有自动测量结果与手动收集的测量结果在统计学上一致。这种方法将对动物遗传学家、科学家和从业者有吸引力，可用于获得客观的角度测量值，这些测量值可纳入后备母猪替换标准，以优化母猪繁殖单元。
