A dataset of target detection for dairy cow body condition scores

<p indent="0mm">Body condition score is an important indicator for evaluating the productivity and health of dairy cows. With the development of modern farms, intelligent detection technologies are now applied to the precision breeding of dairy cows. This study aims to create a dataset of target detection for dairy cow body condition scores. dairy cow body condition score target detection dataset to support the automation of dairy cow health monitoring in the agricultural field. The dataset consists of 5 types of images with body condition scores of 3.25, 3.5, 3.75, 4.0, and 4.25 for multiple cow images, each labeled with a body condition score area. Body condition score is an important measure of cow health, which is of great significance to accurate and efficient evaluation. The images were collected from large-scale farms, with each cow scored by a team of professional veterinarians. The scores were then used to generate annotation information, which includes the bounding box coordinates of the body condition score area and the corresponding scoring label. To enhance the diversity and robustness of the dataset, we introduced a number of challenging factors into the dataset, such as different lighting conditions, changes in cow posture, and background clutter. This helps to ensure that the trained object detection model exhibits good generalization performance in real-world scenarios. The development of this dataset can provide a powerful tool for cow health monitoring in the agricultural sector, which is expected to advancements in related technologies, improving the efficiency and accuracy of cow management. In addition, the dataset serves as a valuable resource for researchers to delve deeper into the connection between body condition and health in dairy cows, supporting future innovations in agricultural technology.