Marble-Classification-Using-Deep-Learning-

📌 It is a deep learning project that can classify 25 different marble types. It is integrated into the mobile application. The article and the data set, which includes the steps from start to finish, are shared as open source, including the codes.

📌 You can find all the details in the article.

📌 Team Members: Mert İhsan US, Metehan UZUN, Göksu KARACA, Bengisu DOĞANAY

📌 Supervisor: Doç. Dr. Kasım Murat KARAKAYA

🔻 Don't forget to give reference if you use the project, thanks. 🔻

Project Link : https://play.google.com/store/apps/details?id=app.ij.mlwithtensorflowlite

Data set Link : https://drive.google.com/drive/folders/1rFN_iSeAKstfWonKxtRcye67UWXFBDrl?usp=sharing

Screenshots of Project

Abstract

In some cases, marble types can be hard to distinguish by consumers and producers. With the system we will develop in our project, the difficulty of distinguishing marble types, which are widely used in the construction industry, will be eliminated. Our project will provide convenience both to consumers and manufacturers within the sector.

In this project, after teaching the machine via machine learning, the types of marble with image recognition with a mobile application by uploading any marble photo taken with the phone's camera to the system. The type and information of the marble will be transferred to the user.

A consumer who is not in the construction industry and does not have detailed information about marble pays attention to the color and pattern of the marble when purchasing marble. It is quite difficult to distinguish this difference between the marbles with the human eye. This is also a problem for the manufacturers. Marbles, prepared during mass production, are controlled by the human eyes. After a while, the human eye gets tired. So, small details can be overlooked. At the beginning of this project, it is aimed to solve these two problems and develop an application that is useful for both the manufacturer and the consumer side.

Keywords: Marble, Image Processing, Deep Learning, Computer Engineering, Artificial Intelligence, Machine Learning, Mobile Application, Android Studio, TensorFlow Lite

Acknowledgements

📌 We would like to express my deepest appreciation to all those who provided us the possibility to complete this project. A special gratitude we would like to give to our senior project supervisor, Mr. Kasım Murat KARAKAYA