Regression Analysis for Dairy Cattle Body Condition Scoring Based on Dorsal Images

Body condition scoring (BCS) provides an objective assessment of the amount of subcutaneous fat appositions and energy storage of the dairy cattle, which has become a powerful tool for dairy industry management. Traditional BCS is estimated by technicians manually method involving visual and tactile aspects, which is high-cost and subjective. Hence, the, accurate and efficient BCS automatic evaluation technology is studied. In this paper, we proposed an effective dairy cattle body condition scoring method based on cow’s dorsal 2D digital images. A bounding rectangle normalization method is used to extract cow’s contour information and distance vectors are constructed to describe the shape. Then, six regression methods are discussed for BCS regression modeling. Experiment on a benchmark dataset demonstrated that elastic net obtained the best accuracy on the BCS task.

[1]  P. E. Wagner,et al.  A Dairy Cow Body Condition Scoring System and Its Relationship to Selected Production Characteristics , 1982 .

[2]  Jiang Wei,et al.  Study on application of image recognition technique in the cow body condition score , 2011, 2011 IEEE 13th International Conference on Communication Technology.

[3]  D T Galligan,et al.  Principal descriptors of body condition score in Holstein cows. , 1994, Journal of dairy science.

[4]  P. Faverdin,et al.  Rear shape in 3 dimensions summarized by principal component analysis is a good predictor of body condition score in Holstein dairy cows. , 2015, Journal of dairy science.

[5]  Hichem Sahbi,et al.  Kernel PCA for similarity invariant shape recognition , 2007, Neurocomputing.

[6]  Alexander J. Smola,et al.  Support Vector Regression Machines , 1996, NIPS.

[7]  Farid Kianifard.  Applied Linear Regression Models , 1984 .

[8]  Wolfgang Junge,et al.  Estimation of backfat thickness using extracted traits from an automatic 3D optical system in lactating Holstein-Friesian cows , 2014 .

[9]  R. Snee,et al.  Ridge Regression in Practice , 1975 .

[10]  Peter Bühlmann.  Regression shrinkage and selection via the Lasso: a retrospective (Robert Tibshirani): Comments on the presentation , 2011 .

[11]  J M Bewley,et al.  Potential for estimation of body condition scores in dairy cattle from digital images. , 2008, Journal of dairy science.

[12]  H. Zou,et al.  Regularization and variable selection via the elastic net , 2005 .

[13]  S. Ivemeyer,et al.  Comparison of a classical with a highly formularized body condition scoring system for dairy cattle. , 2014, Animal : an international journal of animal bioscience.

[14]  Zhengyou Zhang,et al.  Comparison between geometry-based and Gabor-wavelets-based facial expression recognition using multi-layer perceptron , 1998, Proceedings Third IEEE International Conference on Automatic Face and Gesture Recognition.

[15]  Thomas B Farver,et al.  A Body Condition Scoring Chart for Holstein Dairy Cows , 1989 .

[16]  R. Tibshirani,et al.  Regression shrinkage and selection via the lasso: a retrospective , 2011 .

[17]  G. C. Guarnera,et al.  Objective estimation of body condition score by modeling cow body shape from digital images. , 2011, Journal of dairy science.