International Conference on Information and Knowledge Technology

فارسی

Home / سیزدهمین کنفرانس بین المللی فناوری اطلاعات و دانش

Multi-label Classification of Steel Surface Defects Using Transfer Learning and Vision Transformer

Authors :

Amirhossein Komijani¹ Farzaneh Vafaeinezhad² Javad Khoramdel³ Yasamin Borhani⁴ Esmaeil Najafi⁵

1- دانشگاه صنعتی خواجه نصیرالدین طوسی 2- دانشگاه صنعتی خواجه نصیرالدین طوسی 3- دانشگاه تربیت مدرس 4- دانشگاه صنعتی خواجه نصیرالدین طوسی 5- دانشگاه صنعتی خواجه نصیرالدین طوسی

Keywords :

Computer vision،multi-label classification،steel defect detection،deep learning

Abstract :

Steel is the most widely used metal in various industries, for instance, automotive, buildings, packaging, etc. Defect detection on steel surfaces is crucial for steel production companies. This paper proposes a multi-label classification algorithm based on deep learning methods. Several models were trained based on the Severstal dataset, namely MobileNet- V2, Xception, DenseNet121, ViT, and ResNet-50 with transfer learning. In addition, a base model is trained from scratch based on convolutional layers such as ResNet. Moreover, a ViT model is constructed according to the attention and transformer layers concept. At first, 91% for the weighted F1 score is obtained. By exploiting the weighted loss method, the weighted F1 score increased to 92%.

List of archived papers

بررسی تأثیر استقرار استاندارد COBIT در افزایش بهره وری سازمان‌ها (مطالعه موردی: شعب نمایندگی‌های همراه اول، ایرانسل، رایتل)

دکتر محمد ابراهیم سمیع - ساره رحمانیان محمد ابراهیم سمیع - ساره رحمانیان -

Optimal control of robotic hand for rehabilitation using fractional order systems and EEG signal processing

Mehran Safari Dehnavi - Vahid Safari Dehnavi - Masoud Shafiee

Target-driven Navigation of a Mobile Robot using an End-to-end Deep Learning Approach

Mohammad Matin Hosni - Ali Kheiri - Esmaeil Najafi

Presenting an Edge-based Air Quality Management System for Smart City Scenarios

Tina Samizadeh Nikoui - Ali Balador - Amir Masoud Rahmani - Hooman Tabarsaied

The risk prediction of heart disease by using neuro-fuzzy and improved GOA

Vahid Safari Dehnavi - Masoud Shafiee

A Topic Based Method to Classify the Question Clarity in CQA Networks

Alireza Khabbazan - Dr Ahmad Ali Abin

پیش‌بینی بازار فارکس با استفاده از نمودار شمعی و شبکه‌ی عصبی GRU

محمدرضا نوروزی - مریم مومنی

SecVanet: provably secure authentication protocol for sending emergency events in VANET

Seyed Amir Mousavi - Mohammad Sadeq Sirjani - Seyyed Javad Bozorg zadeh Razavi - Morteza Nikooghadam

ML-based Optical Fibre Fault Detection in Smart Surveillance and Traffic Systems

Rushil Patel - Sana Narmawala - Nikunjkumar Mahida - Rajesh Gupta - Sudeep Tanwar - Hossein Shahinzadeh

قطعه بندی خودکار توده کلیه در تصاویر توموگرافی کامپیوتری با استفاده از هم‌افزایی شبکه عصبی عمیق U-Net و الگوریتم فراابتکاری نهنگ

علی خلیلی - محمد مصلح - محمد خیراندیش

more

Samin Hamayesh - Version 42.5.2