Automated defect detection in perfume bottle packaging using machine vision approach for improved quality control

Rafan, Nur Aidawaty and Ismail, Ahmad Najmuddeen and Norizan, Nor Affendy (2025) Automated defect detection in perfume bottle packaging using machine vision approach for improved quality control. Multidisciplinary Science Journal, 7 (8). pp. 1-8. ISSN 2675-1240

Text 0079328032025353151722.pdf Download (675kB)

Abstract

In the cosmetic industry, the quality of perfume bottle packaging was essential for both aesthetic appeal and functionality, as plastic bottles were susceptible to defects that could compromise their attractiveness and usability. Traditionally, these bottles were subjected to manual inspection, a method that was both time-consuming and frequently inaccurate. However, the bottle is manufactured with molding techniques prone to have surface defect that impact structural integrity and appearance of the bottles. Machine vision technology is one of the approaches to improve defect identification with better accuracy. This paper aims to evaluate the designated automated machine vision system to detect defects in plastic perfume bottles with greater efficiency. The system utilized a USB camera and MATLAB software to capture and analyze images of the bottles, employing a black and white detection method to identify imperfections. It integrated a laptop, Arduino UNO, LED light bar, USB camera, MG995 servomotor, and power bank to facilitate the automated sorting of defective bottles. It is determined that threshold 0.2 is the most suitable for the accuracy testing. The automated bottle defect detector manages to obtain an accuracy of more than 95% for every batch inspected. Optimizing the timing of image capture and rotation, incorporating advanced image processing algorithms, and integrating automated systems for bottle placement and removal can significantly improve performance. The overall reliability and accuracy of automatic inspection systems can be further enhanced. This innovative approach provided notable advantages, including more reliable and scalable inspections, which contributed to superior product quality, heightened customer satisfaction, and potential cost savings.

Item Type:	Article
Uncontrolled Keywords:	Automated inspection, Defect detection, Machine vision, MATLAB, Quality control
Divisions:	Faculty Of Industrial And Manufacturing Technology And Engineering
Depositing User:	Norfaradilla Idayu Ab. Ghafar
Date Deposited:	13 Apr 2026 07:38
Last Modified:	13 Apr 2026 07:38
URI:	http://eprints.utem.edu.my/id/eprint/29636
Statistic Details:	View Download Statistic

Actions (login required)

View Item