Defect detection in production of corrugated carton using image processinga approach

Chai, Yu Ki (2024) Defect detection in production of corrugated carton using image processinga approach. Project Report. Universiti Teknikal Malaysia Melaka, Melaka, Malaysia. (Submitted)

Text (Full text) Defect detection in production of corrugated carton using image processinga approach.pdf - Submitted Version Download (3MB)

Abstract

In the manufacturing industry, consistent production of high-quality goods is critical to ensuring customer satisfaction and maintaining a competitive advantage. A corrugated carton manufacturer, is facing an issues during the inspection process, such as a failure to detect defects, a lack of accuracy, and inefficiencies in defect identification. The use of manual inspection methods exacerbates these issues, resulting in longer processing times and an increase in customer complaints. Recognizing the critical need for a solution, this project proposes an image processing approach to address these issues. The project's primary goals are threefold: first, to identify the most critical defects in corrugated carton production; second, to implement an image processing method specifically designed for defect detection; and third, to evaluate the performance of the image processing approach in terms of processing time and accuracy. The project involves five key steps: identifying critical defects, selecting an image processing method, adopting a MATLAB algorithm, debugging MATLAB code, and analyzing automatic inspection. MATLAB Simulink software is used to generate the algorithm and Graphical User Interface (GUI). A total of 30 samples were inspected, achieving 97% accuracy and reducing processing time from 14.49s to 9.29s. Controlling illumination in the inspection environment is crucial for improving quality. Future enhancements may include upgrading the system to detect 3D defects for better performance and utility.

Actions (login required)

View Item