Optimization of light emission for product quality inspection by using machine vision

Zainon, Maslan and Zamri, Ruzaidi and Md Fauadi, Muhammad Hafidz Fazli and Paiman, Nur Aisyah and Zulhasny, Luqman (2024) Optimization of light emission for product quality inspection by using machine vision. Multidisciplinary Science Journal, 7 (7). ISSN 2675-1240

Text 3031 Download (21kB)

Abstract

Automated Visual Inspection (AVI) systems are essential for tasks such as piece counting, position recognition, and form identification, which are crucial for sorting, measurement, testing, and inspection. Despite these benefits, AVI systems are highly sensitive to image quality, which is heavily influenced by lighting conditions. Poor light quality can impair image clarity, resulting in measurement inaccuracies and reducing inspection reliability. This research aims to evaluate and optimize the performance of AVI systems under three different light sources: LED, fluorescent, and LED + fluorescent, focusing on enhancing image clarity and accuracy. The AVI system’s performance was assessed based on lux level standards as per the Indian Standard (IS 6665), utilizing correlation coefficient analysis to evaluate the linear relationship between light intensity and image quality. Additionally, image segmentation techniques were applied to analyze the distribution of bright and dark pixels in order to quantify visual contrast and object distinction. The correlation coefficient model indicated that the “LED + fluorescent” condition performed best, but histogram analysis of pixel intensity revealed that the LED-only setup produced an optimal, well-balanced histogram with distinct intensity peaks for object and background pixels. This balance enhances object detection and background differentiation, leading to improved inspection precision. Consequently, the “LED” condition is considered the best for achieving clear and accurate imaging results in AVI systems. This research highlights the critical role of optimized lighting in AVI systems, providing valuable insights into how light source selection can significantly influence imaging outcomes. This research contributes to the understanding of how different light sources affect AVI system performance and offers practical guidelines for enhancing image quality and reducing errors in AVI-based quality control.

Item Type:	Article
Uncontrolled Keywords:	Automated visual inspection (AVI), In-sight explorer, Light emission, Otsu’s method
Divisions:	Faculty Of Industrial And Manufacturing Technology And Engineering
Depositing User:	Sabariah Ismail
Date Deposited:	23 May 2025 16:24
Last Modified:	23 May 2025 16:24
URI:	http://eprints.utem.edu.my/id/eprint/28694
Statistic Details:	View Download Statistic

Actions (login required)

View Item