Image processing and segmentation for fringe projection profilometry image

• Main Author: Tan, Jeremy Kiang Hwee
• Other Authors: Qian Kemao
• Format: Final Year Project
• Language: English
• Published: Nanyang Technological University 2024
• Subjects: Computer and Information Science; Image segmentation; Computer vision
• Online Access: https://hdl.handle.net/10356/175169
• Institution: Nanyang Technological University

Fringe projection is a widely used technique in 3D scanning. However, one of the critical challenges in fringe projection is the accurate segmentation of fringe patterns from captured images. This project aims to explore and develop various image segmentation techniques to extract fringe patterns from fringe projection profilometry images effectively. Objectives To develop and compare different image segmentation methods for extracting fringe patterns from fringe projection images. To assess the accuracy and efficiency of each segmentation method in real-world scenarios. To evaluate the effectiveness of the deep learning approaches in improving fringe pattern extraction. Purpose of the project The primary objective of this project is to develop and evaluate image processing and segmentation techniques specifically tailored for fringe projection profilometry images. Fringe projection images are commonly used in 3D scanning. Accurate segmentation of fringe projection profilometry images is crucial for extracting meaningful information and improving the quality of subsequent image processing tasks. Image segmentation using computer vision methods and deep learning methods is a widely research topic. The challenge of image processing and segmentation of fringe projection profilometry images has not been fully explored yet. This project aims to implement various techniques of image processing and segmentation of fringe projection profilometry images in aiding the extraction of information for 3D scanning.