This Notebook contains the project I developed as part of the Engineering Design-2 course at Kocaeli University. The project focuses on defect detection on metal surfaces and aims to provide solutions with using YOLOv9.
This Notebook is a implementation of a tutorial, which is written by Piotr Skalski & James Gallagher.
The YOLOv9 algorithm, short for "You Only Look Once version 9," is a deep learning model used for real-time object detection and classification. In the context of defect detection on metal surfaces, YOLOv9 is trained to identify and locate various types of defects, such as scratches, dents, or cracks, on metal surfaces.
The purpose of utilizing YOLOv9 for defect detection on metal surfaces is to enhance quality control processes in industrial manufacturing. By automating the detection of defects, manufacturers can identify flaws in metal components more efficiently and accurately. This allows for timely intervention to address issues, thereby improving product quality, reducing production costs, and ensuring compliance with quality standards.
The project uses the following technologies and tools:
- Programming Language: Python
- Model & Weight: YOLOv9 & GELAN-C
- Dataset Preparation: Roboflow
To run the project on your local machine, follow these steps:
- Clone this repository:
git clone https://github.com/fserdeniz/Metal-Surface-Defect-Detection-with-YOLOv9.git
I also extend my gratitude to my project advisor Ayhan Küçükmanisa and my friend İlhan Kaan Yazıcıoğlu.
This project is licensed under the Apache 2.0 License - see the LICENSE file for details.