Sheep Pregnancy Classification Dataset

Published: 2 July 2025| Version 1 | DOI: 10.17632/vm422yf2cr.1
Contributors:
Christian Lazo, Claudia Letelier

Description

The Sheep Pregnancy Classification Dataset (SPCD) consists of images of dairy sheep classified according to their gestational status and stage of pregnancy. The photographs were captured during the winter season in southern Chile, under natural lighting conditions. Notably, the sheep were not being milked; instead, they were individually fed in the milking area as part of routine nutritional management within a semi-stabled housing system. The photographs were taken on different days using a smartphone camera, reflecting real farm conditions and capturing natural variations in posture, lighting, and the individual appearance of the sheep during regular feeding sessions. Pregnancy status was independently confirmed through ultrasound diagnosis and subsequently validated through lambing, ensuring a reliable ground truth for classification tasks.

Files

Steps to reproduce

A group of dairy sheep, both pregnant and non-pregnant, was monitored during routine feeding sessions in a milking barn in southern Chile. Pregnant ewes were followed through mid and late gestation. Photographs were taken under natural lighting using a smartphone camera. Gestational status was confirmed via ultrasound and later validated through lambing.

Institutions

  • Universidad Austral de Chile

Categories

Artificial Intelligence, Pregnancy, Sheep, Pattern Recognition, Precision Agriculture

Licence