Glass bottle defect detection by using image processing technique

Noh, Ahmad Aizat (2015) Glass bottle defect detection by using image processing technique. Project Report. Universiti Teknikal Malaysia Melaka, Melaka, Malaysia. (Submitted)

	Text (24 Pages) Glass Bottle Defect Detection By Using Image Processing Technique.pdf - Submitted Version Download (457kB)
	Text (Full Text) Glass Bottle Defect Detection by Using Image Processing Technique.pdf - Submitted Version Restricted to Repository staff only Download (1MB)

Abstract

Machine vision is the technology and technique that has been implemented in the industrial sectors all over the world recently. The technology is based on the image where it can be used for real time capture image application such as in inspection, process control or robot guidance in industry which allowing the process run faster, high accuracy, good repetitive motion and safe compared to human works output. The idea of applying the vision technology on inspecting the glass bottle is to replace the current technology that used the photoelectric detecting system with touching and rolling mechanism. The aim of this project was to build a vision system that can detect the hole and the body roundness of the bottle. The methods used to identify and classified the object was based on the former researcher works. From the journal mapping, the edge detection method was applied to inspect the sample. The technology consists of software and hardware, the software used for developing the algorithm was chosen based on the author’s knowledge in programming. The algorithm has been developed in MatLab software. By using the algorithm, the parameter and measurement analysis have been performed to the sample. From this experiment, all of the defect samples have been successfully detect. The hardware had identified the defect product and produced sound as the signal to detect defection. For future development, it was suggested to create jig to hold the sample, adjust the lighting position and improve the type of defect detection.

Actions (login required)

View Item