Relevant Algorithms Study And Software System Design For Micro-CT

Computed Tomography (CT) is a medical imaging method, which can obtain the structure information of the detected object based on the fact that the absorption rate of X-ray is different in different materials. It has been widely used in many domains, such as industry, safety and medicine. In particular, CT has become an indispensable diagnostic tool in the application of medicine. However, the common medical CT is not suitable for small animal imaging due to the high cost, high radiation and low spatial resolution. To handle this problem, Micro-CT is developed. This thesis concentrates on the image stitching algorithm and three-dimensional reconstruction algorithm for Micro-CT. On the basis of the proposed algorithms, the corresponding software platform is developed for Micro-CT system.Data acquired by the Micro-CT system need to be processed before reconstruction. Firstly, we need to remove the bad points, bad lines and noise which are mainly caused by the equipment or operation error and the environment influence. Secondly, the positions of the emitter, object and detector are determined by the geometric correction process. Finally, the reconstruction process is performed on the aforementioned processed data. Because of the limitation of the detector area, the object is usually divided into several parts to obtain complete data. For this purpose, an template matching based image stitching method is proposed in this thesis to further process the data after preprocessing and geometric correction. Furthermore, FDK algorithm is used to reconstruct the three-dimensional image of the object. A series of Experiments are conducted to demonstrate the accuracy and feasibility of the proposed image stitching and reconstruction algorithm.Based on the aforementioned work, a software platform is developed for Micro-CT system. It includes the acquisition controlling module, data preprocessing module, 3D reconstruction module and data post-processing. Because of the modularization design and visualization interface, the software platform provides a powerful and user-friendly platform for the researchers of the medical imaging field.