Metal surface defect detection and quality evaluation using deep learning
Author(s):
Supreetha M S, Hrushika M, Surabhi N, Pruthvi K V, Nandini B M
Keywords:
Deep learning, Convolutional Neural Network, Random Forest, K Nearest Neighbour, Metal defect detection.
Abstract
Detecting defects on metal surface is vital for businesses to preserve quality measures of the item and to help excess in generation. With this work, we put forward three machine learning (ML) classifiers- Convolutional Neural Network, Random Forest, K Nearest Neighbour to distinguish, detect and classify the deformity and defect within the dataset. Firstly, information is pre-processed to format images. At that point the models are utilized to train defect detection classification assignment with finest combination of weights and bias to ML calculation. Besides, quality evaluation is done among the three models with the assistance of diverse criteria from classification report.
Article Details
Unique Paper ID: 155883
Publication Volume & Issue: Volume 9, Issue 2
Page(s): 270 - 276
Article Preview & Download
Share This Article
Join our RMS
Conference Alert
NCSEM 2024
National Conference on Sustainable Engineering and Management - 2024