Development Of Pipeline Corrosion Inspection System Using Machine Vision
These days, utilization of camera as an inspection tool has been expanded. The flexible function of camera is adequate to obtain different kind of information. In Cawley (2001) review on NDT that was presented in 2001, Radiography, Ultrasonic, Eddy Current, Magnetic Particle, and Penetrant Testing w...
Saved in:
| Main Author: | |
|---|---|
| Format: | Thesis |
| Language: | English English |
| Published: |
2016
|
| Subjects: | |
| Online Access: | http://eprints.utem.edu.my/id/eprint/20463/1/Development%20Of%20Pipeline%20Corrosion%20Inspection%20System%20Using%20Machine%20Vision.pdf http://eprints.utem.edu.my/id/eprint/20463/2/Development%20Of%20Pipeline%20Corrosion%20Inspection%20System%20Using%20Machine%20Vision.pdf http://eprints.utem.edu.my/id/eprint/20463/ https://plh.utem.edu.my/cgi-bin/koha/opac-detail.pl?biblionumber=105854 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Universiti Teknikal Malaysia Melaka |
| Language: | English English |
| Summary: | These days, utilization of camera as an inspection tool has been expanded. The flexible function of camera is adequate to obtain different kind of information. In Cawley (2001) review on NDT that was presented in 2001, Radiography, Ultrasonic, Eddy Current, Magnetic Particle, and Penetrant Testing were the top five techniques dominating the NDT market yet Visual Inspection is the most widely applied. Even though the popularity of visual inspection is higher compared to other NDT method, but due to the reliability issues it is often used together with other methods. This research work is focusing on developing a robust corrosion inspection system based on vision sensor that is able to accurately detect and classify corrosion based on the appearance features. By installing at an early stage, inspection system would be able to gather data and at the same time identify and analyse the collected data. Through the results, the analysed data is able to classify the corrosion type based on appearance. From the research work, the method of using image enhancement filters to improve accuracy of vision corrosion inspection system is identified. The detection of each macroscopic surface corrosion types; galvanic; crevice; erosion; pitting and exfoliation using vision inspection able to achieve 79% accuracy using the simulated dataset. The new method of corrosion inspection operation which able to generate prevention plan has qualified the Vision Corrosion Inspection System to be used during preliminary inspection. It is expected that the Vision Corrosion Inspection System can improve vision inspection as the pioneer in NDT method for corrosion inspection. In addition, framework of the developed Vision Corrosion Inspection system is applicable for other applications of vision inspection whereby it can be applied for other inspection process or extending its application to other problems. |
|---|