Real Time Tray-level Egg Fertility Detection Using Deep Learning

Project Code :TCMAPY2215

Objective

The primary objective of this project is to develop a deep learning-based solution that leverages image processing techniques to detect egg fertility. By employing advanced object detection models, such as YOLOv11, YOLOv12, and YOLOv26, the project aims to automate the process of classifying eggs into two categories: "Fertile" and "Infertile." This solution seeks to reduce the need for manual inspection, thereby improving efficiency and accuracy in egg sorting.Additionally, the project aims to design a user-friendly web application built on the Flask framework, where users can seamlessly upload egg images and receive fertility predictions in real-time. The goal is to ensure that the model performs with high accuracy and minimal errors, specifically focusing on reducing false positives and false negatives.Furthermore, the project demonstrates how deep learning and object detection technologies can play a significant role in agricultural automation, particularly within the poultry industry. By creating a scalable system, the solution is designed to support industrial-scale operations, enabling the classification of large volumes of eggs with ease. Ultimately, the project provides valuable insights into the benefits of automation in agriculture, such as cost reduction, less reliance on manual labor, and enhanced operational efficiency

Abstract

This project focuses on the development of a deep learning-based system for the detection of egg fertility at tray-level. The primary goal is to classify eggs as either "Fertile" or "Infertile" using a state-of-the-art object detection model, YOLOv12. The dataset used for this task is sourced from Roboflow and contains images of eggs that are labeled into two distinct categories. The project leverages the capabilities of the YOLO series (v11, v12, and v26) to perform object detection with high accuracy. The solution is implemented using a Flask backend with a front-end interface designed with HTML, CSS, and JavaScript. The model processes images and delivers real-time predictions of egg fertility, assisting in automated sorting and improving efficiency in the agricultural sector. This approach not only reduces the manual effort required for fertility detection but also offers scalable and efficient support for egg classification in commercial poultry farming.

Keywords: egg fertility detection, YOLO, deep learning, object detection, YOLOv12, Flask, image classification, fertility prediction, Roboflow, egg candling.

NOTE: Without the concern of our team, please don't submit to the college. This Abstract varies based on student requirements.