Classifying cow stall numbers using YOLO

TL;DR

This paper presents a new dataset and an improved YOLO-based method for accurately classifying cow stall numbers from images, achieving over 95% accuracy in a practical agricultural setting.

Contribution

The introduction of the CowStallNumbers dataset and the application of fine-tuned YOLO with data augmentation for cow stall number classification.

Findings

Achieved 95.4% accuracy in stall number recognition.

Created a new dataset with over 1300 images for this task.

Demonstrated effectiveness of data augmentation and fine-tuning.

Abstract

This paper introduces the CowStallNumbers dataset, a collection of images extracted from videos focusing on cow teats, designed to advance the field of cow stall number detection. The dataset comprises 1042 training images and 261 test images, featuring stall numbers ranging from 0 to 60. To enhance the dataset, we performed fine-tuning on a YOLO model and applied data augmentation techniques, including random crop, center crop, and random rotation. The experimental outcomes demonstrate a notable 95.4\% accuracy in recognizing stall numbers.