Machine Vision-Based Surface Defect Detection of Metal Components Using Digital Image Processing

Metal parts can get scratches, cracks and other defects on their surface. These defects spoil the quality and reduce the life of the part. Checking them by eye is slow and tiring, and two people may give two different answers. In this paper we plan a simple machine vision system that finds and sorts surface defects using digital image processing. The system takes an image, converts it to grayscale, removes noise, improves contrast, separates the defect area, cleans it with morphology, measures some features and then decides the defect type. We plan to test it on the NEU surface defect database, which has 1,800 images of six steel defect types. We want to keep the method light and low-cost, so it does not need a heavy deep-learning model. Accuracy, precision, recall, F1score and a confusion matrix will be used to check the performance. Experimental results are not included yet and will be added after the experiments are done.

Authors

Publication Details

Journal
Zenodo (CERN European Organization for Nuclear Research)
Published
2026-11-19
DOI
https://doi.org/10.5281/zenodo.24371608
Primary Topic
Industrial Vision Systems and Defect Detection
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Machine Vision-Based Surface Defect Detection of Metal Components Using Digital Image Processing

Sarthak Erande
Zenodo (CERN European Organization for Nuclear Research)
Industrial Vision Systems and Defect Detection
article

Machine Vision-Based Surface Defect Detection of Metal Components Using Digital Image Processing

Sarthak Erande
article en

Abstract

Metal parts can get scratches, cracks and other defects on their surface. These defects spoil the quality and reduce the life of the part. Checking them by eye is slow and tiring, and two people may give two different answers. In this paper we plan a simple machine vision system that finds and sorts surface defects using digital image processing. The system takes an image, converts it to grayscale, removes noise, improves contrast, separates the defect area, cleans it with morphology, measures some features and then decides the defect type. We plan to test it on the NEU surface defect database, which has 1,800 images of six steel defect types. We want to keep the method light and low-cost, so it does not need a heavy deep-learning model. Accuracy, precision, recall, F1score and a confusion matrix will be used to check the performance. Experimental results are not included yet and will be added after the experiments are done.

Zenodo (CERN European Organization for Nuclear Research)
Industrial Vision Systems and Defect Detection
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon