2021/02/25 by Clément Allain, Anaïs Caillot, Allain, Clément +15
Agricultural and Biological Sciences · Biochemistry, Genetics and Molecular Biology · #Area #Cow #Effects of Environmental Stressors on Livestock #Estimation du poids #Genetic and phenotypic traits in livestock #Scanner #Surface #Three-dimensional #Tridimensionnel #Vache #Volume #Weight estimation
paper · doi:10.15454/21gn-kn32
openalex publication_date 2021/02/25 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/29
The development of 3D imaging is a solution for estimating morphological parameters and growth development in dairy cattle but it also offers opportunities to estimate original indicators such as volume, body surface area and body weight. The objective of the Morpho3D project was to develop, test and validate a device offering the possibility to record and analyze complete 3D shapes of dairy cattle. To evaluate the performance of this tool, two trials were carried out on Holstein cows, one in 2017 on 30 cows and one in 2018 on 64 cows. The precision of linear measurements (height at the withers (HG), chest circumference (TP), chest depth (PP), hip width (LH), backside width (LF) and ischial width (LI)), volume and body surface area were first tested on inert forms (plastic cow and cylinder). The differences observed between manual reference measurements and estimates on the 3D image were less than 1% for area and volume and ranged from 1.3% to 11.2 % for linear measurements. For the repeatability and reproducibility of the method for estimating volume and area in living animals, the coefficients of variation (CVs) were 0.17 and 3.12 % respectively. For linear measurements, the CVs ranged from 0.26 to 9.81 % for repeatability and from 0.94 to 7.34 % for reproducibility. Four models for estimating the liveweight of cows were established. The coefficients of determination ranged from 0.82 to 0.93 % with weight prediction errors around 3 % (20 kg) as function of the volume and 4 % (29 kg) using the area. The device and the method, evaluated and validated in the present study, offer possibilities of using new indicators such as volume and body surface in precision breeding.