Three-dimensional images methodology to morphometric evaluation of Mangalarga Marchador horses

• Ano de defesa: 2025
• Autor(a) principal: Alves, Tiago Tolêdo Bittencourt e
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: eng
• Instituição de defesa: Universidade Federal de Viçosa; Zootecnia
• Palavras-chave em Português: Mangalarga Marchador (Cavalo) - Morfologia; Mangalarga Marchador (Cavalo) - Pesos e medidas, Imagens tridimensionais; Mangalarga Marchador (Cavalo) - Registros de desempenho; CIENCIAS AGRARIAS::ZOOTECNIA::PRODUCAO ANIMAL
• Link de acesso: https://locus.ufv.br/handle/123456789/34180; https://doi.org/10.47328/ufvbbt.2025.255

In equine science, we correlate biometric parameters with locomotion characteristics and sporting performance. Those are some tools that assist in the evaluation and selection breeding process. Traditionally, morphometric measurements are obtained manually, requiring evaluators and need to have contact with the animal. That methodology makes the process laborious, slow, and stressful. Furthermore, subjectivity may occur among evaluators, compromising the accuracy of data collection. The objective of this research was to establish a non-invasive, accurate, and easy-to-use methodology for reconstructing the 3D models and carrying out morphometric evaluation of horses. Fifty-two (52) Mangalarga Marchador horses were used to construct the 3D models, and the following ten measurements were taken: withers height (WH); hip height (HH); head length (HL); neck length (NL); back-loin length (BLL); croup length (CL); shoulder length (SL); body length (BL); head width (HW); and croup width (CW). Conventional and digital methods were used to take morphometric measures. Statistical analysis (Pearson Correlation ad linear regression) are used to establish the correlation between the two methodologies to verify the accuracy and precision. The results showed high precision, with R² of 98,78% when taking the overall measurements. Some animal parts reached a moderate correlation (HaW, HaC, SL, BL, CW), and for the others, the precision was low. The methodology is robust and reliable but needs some improvements for better accuracy. Keywords: Horse conformation; Digital; Animal production; Breed characterization.