(For USM Staff/Student Only)

EngLib USM > Ω School of Electrical & Electronic Engineering >

Development of machine vision inspection system using line scan camera for label printing

Development of machine vision inspection system using line scan camera for label printing / Muhamad Faris Che Aminudin
Teknologi penglihatan mesin masih berkembang dan bertambah baik setiap tahun. Penglihatan mesin banyak digunakan di dalam pelbagai kegunaan seperti di dalam sektor pembuatan, pemantauan trafik, dan pembedahan. Teknologi ini digunakan untuk membantu dan memudahkan manusia dari segi penglihatan dengan mengunakan kamera untuk melakukan kerja tertentu tanpa mengunakan mata manusia. Projek ini memberi tumpuan kepada mereka bentuk dan membangunkan sistem pemeriksaan penglihatan mesin menggunakan kamera imbasan garis. Terdapat beberapa faktor yang perlu di ambil kira ketika mereka bentuk sistem cetakan berdasarkan penglihatan mesin seperti jenis kamera yang digunakan, teknik pencahayaan, objek untuk diperiksa dan teknik pemprosesan gambar. Di dalam projek ini kamera imbasan garis lurus GigE digunakan untuk menangkap gambar dan gambar tersebut diproses untuk mengesan kecacatan pada percetakan label. Proses yang digunakan untuk mengesan kecacatan adalah dengan mengunakan “Cascade Classifier” kerana ia lebih teguh berbanding kaedah memadankan template. Sistem pemeriksaan merangkumi kamera, tali pinggang penghantar, pengendali mikro, lampu, dan komputer yang diintegrasikan bersama untuk membina sistem pemeriksaan penglihatan mesin. Analisis prestasi dan ketepatan untuk sistem pemeriksaan dibincangkan di dalam hasil kajian. Ketepatan mengunakan “Cascade Classifier” adalah lebih tinggi berbanding dengan menggunakan persamaan gambar. _______________________________________________________________________________________________________ Machine vision technology keeps developing and improving over the year. The vision system is used in wide varieties of applications such as manufacturing, traffic monitoring and surgery. This technology is used to ease and help human in term of vision by using a camera instead of human eyes to do the specific task. This project is focusing on designing and developing line-scan camera based machine vision inspection system. There are several factors that need to be considered when designing a machine vision based inspection system such as the type of camera, lighting technique, object that need to be inspected and image processing technique. In this project the inspection system consists of camera, conveyor belt, microcontroller, light, and computer that are integrated together to build the machine vision inspection system is developed. The GigE line-scan camera is used in this project to capture the image and then the image will be processed to detect defect in label printing. The process of detecting the defect is processed and tested by using cascade classifier and template matching and both method are compared. Template matching that were used in this project is not robust to rotation and scaling that causing lower performance compare to the cascade classifier. The cascade classifiers are more robust in scaling and rotation than normal template matching. The analysis of the inspection system performance and accuracy are discussed in the result section. The cascade classifier show an average accuracy of 93 percent while as the template matching show the accuracy average of 30 percent. Therefore cascade classifier is used in the inspection system because it has higher accuracy and more robust.
Contributor(s):
Muhamad Faris Che Aminudin - Author
Primary Item Type:
Final Year Project
Identifiers:
Accession Number : 875006338
Language:
English
Subject Keywords:
vision technology; manufacturing; surgery
First presented to the public:
6/1/2016
Original Publication Date:
6/5/2018
Previously Published By:
Universiti Sains Malaysia
Place Of Publication:
School of Electrical & Electronic Engineering
Citation:
Extents:
Number of Pages - 102
License Grantor / Date Granted:
  / ( View License )
Date Deposited
2018-06-05 15:16:33.816
Date Last Updated
2019-01-07 11:24:32.9118
Submitter:
Mohd Jasnizam Mohd Salleh

All Versions

Thumbnail Name Version Created Date
Development of machine vision inspection system using line scan camera for label printing1 2018-06-05 15:16:33.816