Automated surface registration of 3D objects.

• 主要作者: Tse, Chung Yan.
• 其他作者: Lee Kijoon
• 格式: Final Year Project
• 語言: English
• 出版: 2009
• 主題: DRNTU::Engineering::Bioengineering
• 在線閱讀: http://hdl.handle.net/10356/16449

Remote Diffuse Optical Tomography (DOT) is an imaging technique that reconstructs the optical properties of biological tissue after illuminating light into them and detecting the scattered light with optical detectors. The aim of this project is to develop a laser scanning system that can be incorporated into remote DOT and obtain 3D camera coordinates of the object scanned. This is done by registering surfaces of 3D objects. This project was proposed to be done through three steps, namely, calibration of CCD camera, scanning of objects with laser and recovery of 3D camera coordinates. Calibration of CCD camera is performed to determine the relationship between 3D world coordinates and 2D image coordinates by finding the intrinsic and extrinsic parameters of the CCD camera. These parameters are useful for analysis of images captured in the later part of the project. In this project, calibration of CCD camera is performed with the help of Camera Calibration Toolbox for Matlab from Caltech [8]. Scanning of objects with laser is performed with laser scanning module suggested by the author and is fully controlled by a laser scanning control software. With the software, laser stripes are scanned on objects and images are captured during each scan with image-acquisition software, PMCapture. Analysis of the images and recovery of 3D camera coordinates are performed with image processing module that is written in Matlab by the author. The method used and results obtained throughout this project are discussed and presented in this report.