Development of 3D defect detection system of automotive parts using image processing approach

Mohd Shaed, Nur Amira Faqihah (2022) Development of 3D defect detection system of automotive parts using image processing approach. Project Report. Melaka, Malaysia, Universiti Teknikal Malaysia Melaka. (Submitted)

	Text (24 Pages) Development of 3D defect detection system of automotive parts using image processing approach.pdf - Submitted Version Download (4MB)
	Text (Full Text) Development of 3D defect detection system of automotive parts using image processing approach.pdf - Submitted Version Restricted to Repository staff only Download (22MB)

Abstract

Image Processing is one of the techniques to extract an important feature from image data, description, interpretation or understanding of the scene can be provided by the camera. The image is converted to an array of small integers, called pixels, representing a physical property such as scene radiance, stored in a digital memory, and processed by a computer or other digital hardware. Manual inspection it has difficulties to find the defects on automotive parts because not enough light and spend too much time to inspect the parts. This study aims to focus on determining the type of 3D defects on automotive parts using image processing techniques. This project aims to design an algorithm for the 3D defect using image processing techniques. Due to the Image Processing Toolbox software is a set of routines that extend the MATLAB programming environment's capabilities. The toolbox supports a wide range of image processing tasks. Then, usi ng confusion matrix for analysing the efficiency of developed an algorithm in terms of accuracy and processing time. The method for this project by using HD CMOS industrial camera and MATLAB for detect the defect on automotive parts. The method for image processing is image pre - processing, detection element and defect detection. The expected result for this project can define 3D defect on automotive parts. It helps to reduce the 3D defect on parts when using image processing techniques. The method and Graphical User Interface (GUI) for this project are generated using the MATLAB Simulink programme. A total of 100 components samples were examined, and the accuracy obtained was 94 percent with a 94.92 second reduction in processing time. To increase the quality of the inspection, the inspection environment must be managed through lighting control. In order to improve performance and make the system more valuable to the organisation, this system's capacity to identify 3D errors may be increased

Actions (login required)

View Item