| Author: | Chen, Hongpeng |
| Title: | PSO-based optimal coverage path planning for surface defect inspection of 3C components with a robotic line scanner |
| Degree: | M.Sc. |
| Year: | 2023 |
| Subject: | Robot vision Robotics Engineering inspection -- Automation Hong Kong Polytechnic University -- Dissertations |
| Department: | Department of Mechanical Engineering |
| Pages: | ix, 96 pages : color illustrations |
| Language: | English |
| Abstract: | Surface defect inspection is a critical process in the quality control and fault diagnosis of computers, communications, and consumer electronics (3C) industry. The demand for automation in defect detection has increased in recent years due to the need for faster, more accurate, and reliable inspection methods compared to traditional approaches. However, the limited field of view of available equipment poses challenges in scanning objects, necessitating the selection of optimal viewpoints and efficient coverage path planning. This dissertation presents a novel approach for an automated scanning system designed to detect surface defects in 3C free-form components. The proposed methodology combines a hybrid region segmentation method, an adaptive algorithm for viewpoint selection, and an optimization method for the global path sequence to improve defect detection efficiency and accuracy. The limited field of view of scanning equipment requires operators, whether human or robotic, to scan objects from multiple viewpoints to ensure comprehensive coverage. To address this challenge, the proposed methodology utilizes a hybrid region segmentation method that integrates geometric and optical parameters. Geometric parameters such as curvature and shape descriptors extract relevant features, while optical parameters like color and intensity variations distinguish defects from the background. The combination of these parameters enhances the accuracy and robustness of the segmentation process, enabling the identification of subtle defects in complex free-form components. Another challenge is the selection of optimal viewpoints for scanning. Efficient coverage path planning plays a crucial role in maximizing defect detection rates while minimizing scanning time. The proposed methodology incorporates an adaptive algorithm for viewpoint selection, considering factors such as defect density, surface complexity, and accessibility. By dynamically adjusting the viewpoint selection based on the workpiece characteristics, the system can adapt to different component types and optimize defect detection rates. Additionally, the proposed methodology focuses on optimizing the global path sequence to maximize scanning efficiency. The sequencing of viewpoints is crucial in reducing travel distances and eliminating unnecessary movements. An optimization method is developed to determine the optimal order of viewpoints, utilizing algorithms such as genetic algorithms or ant colony optimization based on specific system requirements and constraints. By optimizing the path sequence, the proposed methodology significantly enhances scanning efficiency, reducing inspection time and improving overall productivity. To evaluate the performance of the proposed methodology, comprehensive simulation-based and experimental studies were conducted on a variety of free-form workpieces commonly found in the 3C industry. Simulation-based studies involved generating virtual models of components with known defect distributions, allowing for quantitative analysis of detection accuracy and efficiency. Experimental studies utilized a physical scanning system integrated with the proposed methodology, providing real-world data to assess the feasibility and effectiveness of the approach. The results obtained from simulation-based and experimental studies demonstrate the feasibility and effectiveness of the proposed methodology in detecting surface defects in 3C free-form components. The quantitative analysis reveals high detection accuracy and efficiency, showcasing the potential for practical implementation in the industry. Furthermore, a comparative analysis with a state-of-the-art method highlights the superiority of the proposed strategy, emphasizing its advantages in terms of accuracy, efficiency, and adaptability. In conclusion, this dissertation introduces a novel approach for an automated scanning system designed to detect surface defects in 3C free-form components. By combining a hybrid region segmentation method, an adaptive algorithm for viewpoint selection, and an optimization method for the global path sequence, the proposed methodology offers improvements in defect detection efficiency and accuracy. The findings demonstrate its potential for enhancing quality control and fault diagnosis processes in the 3C industry. Future research can explore the integration of deep learning techniques and further optimization methods to further advance the performance of automated scanning systems for surface defect detection in the 3C industry. |
| Rights: | All rights reserved |
| Access: | restricted access |
Files in This Item:
| File | Description | Size | Format | |
|---|---|---|---|---|
| 7467.pdf | For All Users (off-campus access for PolyU Staff & Students only) | 8.61 MB | Adobe PDF | View/Open |
Copyright Undertaking
As a bona fide Library user, I declare that:
- I will abide by the rules and legal ordinances governing copyright regarding the use of the Database.
- I will use the Database for the purpose of my research or private study only and not for circulation or further reproduction or any other purpose.
- I agree to indemnify and hold the University harmless from and against any loss, damage, cost, liability or expenses arising from copyright infringement or unauthorized usage.
By downloading any item(s) listed above, you acknowledge that you have read and understood the copyright undertaking as stated above, and agree to be bound by all of its terms.
Please use this identifier to cite or link to this item:
https://theses.lib.polyu.edu.hk/handle/200/12987