Dissertation > Excellent graduate degree dissertation topics show

Studies on Exact Katsevich Reconstruction Algorithm of Spiral Cone-Beam Industrial CT

Author: NiuXiaoMing
Tutor: ZengLi
School: Chongqing University
Course: Applied Mathematics
Keywords: Industrial spiral cone-beam CT Katsevich exact reconstruction algorithm Adaptive hybrid filter Offset correction Enhanced edge reconstruction
CLC: TP391.41
Type: Master's thesis
Year: 2008
Downloads: 149
Quote: 6
Read: Download Dissertation

Abstract


CT (computed Tomography), namely, computer tomography, which is mathematics, physics, optical, mechanical and computer multidisciplinary combination product. CT their main application areas are divided into medical CT and industrial CT two categories. In nearly a decade, industrial CT scanning method has been the first two-dimensional parallel-beam scanning to today's three-dimensional helical cone-beam scanning. Not only spiral cone-beam industrial CT scanning time is short, high utilization rays, axial distribution, and its three-dimensional scanning track to meet the precise reconstruction of the complete condition has become hot research field of non-destructive testing. Spiral cone-beam CT reconstruction algorithm iterative method and is broadly divided into two categories analytical method. Iterative reconstruction requires multiple iterations, calculates the time-consuming. Analytical method is divided into exact reconstruction and approximate reconstruction, in comparison, the exact reconstruction from cone angle of impact, and the reconstructed image is theoretically no artifacts. Precise reconstruction algorithms are algorithms and Katsevich Grangeat algorithm categories. As the spiral cone-beam algorithm Katsevich algorithm has higher efficiency, better quality of the reconstructed image, etc., at present it is subject to more and more attention. Exact reconstruction requires projection data is complete, without noise. In the actual spiral cone-beam CT in industry, as part of the detector disorders, scattering, crosstalk and other reasons, access to the actual projection data contains a lot of noise. Therefore, exact reconstruction algorithm to improve the projection data or noise, the maximum possible to meet the exact reconstruction condition. In this paper, noisy spiral cone-beam CT reconstruction algorithm Katsevich projection data, to adopt two strategies reduce the impact of noise on the reconstructed image. The first strategy is mainly Katsevich algorithm derivation methods for improvement; second strategy is the use of improved adaptive hybrid filter the projection data noise. Computer simulation experiments confirmed that the first strategy the projection data contains a small amount of abnormal noise and random noise has a better inhibition; second strategy is more suitable for projection data contains a lot of noise. Further, in the actual industrial spiral cone-beam CT, since the ray source and the detector center point level shift highly inconsistent and turn the heart and other reasons, can obtain the actual projection data in different degrees of offset. If the offset of the projection data using the direct image reconstruction, the reconstructed image and the reconstructed image of the ideal large difference. This spiral cone-beam Katsevich by correcting algorithms to adapt to this eccentric CT imaging systems. First, the offset of the projection data is converted to the detector center offsets; in the reconstruction process, according to the center of the detector offset line equation for the filter, the filter before and after the rearrangement and correction, and back-projection process were carried out correction . Among them, the back-projection process, the PI-line traditional iteration Roots unary function into iterative roots, and root-finding method using high iteration optimization was carried out. Computer simulation results show that this method can effectively offset correction reduced projection data offset the impact of the reconstructed image. Meanwhile, the improved method shortens rooting PI line rebuild time. The edge image includes an important feature of the image, the image recognition, segmentation, measurement process has an important role. Industrial CT obtained by the conventional method D object edges are generally divided into two steps: first, using the projection data actually scanned image reconstruction, and the reconstructed three-dimensional image edge extraction. This paper Katsevich spiral cone-beam CT reconstruction formula has been modified, a kind of use of spiral cone-beam CT projection data obtained directly reconstruct three-dimensional object edge enhancement methods. Obtained by the method of three-dimensional edge enhancement, can initially determine the shape and internal structure of the object types, and three-dimensional reconstruction of the time and enhance edges Katsevich spiral cone-beam reconstruction algorithms considerable time, saving time three-dimensional edge extraction.

Related Dissertations

  1. Research on I/Q Imbalance Problems of the OFDM-UWB System,TN919.3
  2. Broadband satellite modulation system synchronization technology research,TN927.2
  3. Rail vehicles welding operating procedures mirroring technology research,U270.6
  4. Research of Image Processing and Defects Recognition Based on X-ray Real-time-image System,TP391.41
  5. Shift Auto-Correlation Frequency Offset Correction of MIMO-OFDM System in Frequency Domain and LDPC New Parameter Decoding,TN919.3
  6. Image sensor image processing technology Algorithm and Hardware Implementation,TP212
  7. The Laser Steady Light and Research That Modulates with a Modulating Device,TN249
  8. Correction of Doppler shift,TN929.5
  9. Based on particle filter for multi -target tracking algorithm,TP391.41
  10. The Correcting Algorithms of Carrier Frequency Offset in Orthogonal Frequency Division Multiplexing (OFDM) Systems,TN919.3
  11. Research on Basic Algorithms of Digital Image Processing and Implementation with FPGA,TP391.41
  12. Research on Facial Feature Extraction and Matching Algorithms for Image Retrieval,TP391.41
  13. Research of High Speed Image Pre-processing System Based on FPGA,TP391.41
  14. Research on Algorithms of 2D Face Template Protection,TP391.41
  15. The Research and Implemention of Image Retrieval Based on User Interested Feature,TP391.41
  16. Research of Image Mosaic Technology,TP391.41
  17. Research and Implementation of Exact String Matchiing Algorithms,TP391.41
  18. Research on the Classification Based on the Reconstruction of Solder Joint,TP391.41
  19. Tongue Feature Extraction and Research of Fusion Classification,TP391.41
  20. The Fatigue State Recognition of the Driver Based on Eye Detection,TP391.41
  21. Research on Infrared Image Simulation for Aerial Objects and Background,TP391.41

CLC: > Industrial Technology > Automation technology,computer technology > Computing technology,computer technology > Computer applications > Information processing (information processing) > Pattern Recognition and devices > Image recognition device
© 2012 www.DissertationTopic.Net  Mobile