Dissertation > Excellent graduate degree dissertation topics show

Research on Correction Method for Cupping Artifacts in the Image of Industrial CT

Author: LiDan
Tutor: WangZuo; DuanXiaoJiao; CaiYuFang
School: Chongqing University
Course: Control Science and Engineering
Keywords: X-ray industrial CT Beam hardening Scatter Crosstalk Cupping artifact
CLC: TP391.41
Type: Master's thesis
Year: 2012
Downloads: 77
Quote: 0
Read: Download Dissertation

Abstract


In the X-ray industrial CT system, the cupping artifact caused by the beamhardening not only affects the quality of the image but also reduces the identificationability of the image and the precision of the density measurement. Therefore, it isespecially important to do the pre-processing operation on the data to correct thecupping artifact in the X-ray industrial CT system.The present thesis introduces the linearized correction algorithm based onpolynomial fitting and its realization detailly. The hardening correction model is alsodesigned. Then this algorithm is used to correct the simulation CT images and actual CTimages respectively. Experimental results show that the algorithm can eliminate thecupping artifacts in CT image and improve the uniformity of the image effectively.However, the algorithm also brings some problems such as the salt-and-pepper noise ofthe corrected image is greater than the original image and the edge artifacts becomeworse. To solve the above problems and obtain better effect, the causes of the problemsare analyzed theoretically. When X-ray interacts with matter, the phenomena of beamhardening and scattering appear. These phenomena together with the crosstalk causedby the scattering in the detector have impacts on the quality of CT images. Therefore, toobtain better correction effect, the influence of scattering and crosstalk must beconsidered simultaneously in the correction of the beam hardening problem.To improve the weakness of using only the linearized correction algorithm, acorrection method based on the combination of software and hardware is proposed. Alead foil sheet is placed between the object to be measured and the collimator to absorbthe scattering ray. Then the linearized correction algorithm is used to correct the beamhardening. To ensure that using one single correction model, better correction effect canbe obtained under different energy, series of hardening curves are obtained by scanningthe hardening correction model with several commonly used X-ray energy. Thecorrection is done by the hardening curves to the CT image obtained under thecorresponding energy. Experiment results show that this correction method eliminatescupping artifacts and reduces the noise at the uniform area of the workpiecesignificantly. Moreover, the uniformity of the image is improved effectively, and boththe detail and edge of the corrected CT image are kept well.To solve the problem of the edge artifact in the CT image, theoretical analysis is done to find the cause of the edge artifact one of which is the crosstalk produced by thescattering of X-ray in the near detector. Mathematic model of the crosstalk in thechannel of the detector is presented. In the following, a correction method based onreducing the coefficient of the crosstalk is proposed. By doing experiments on the mainfactors that influence the crosstalk ratio, the relationship of the crosstalk ratio with theenergy of the incident X-ray and the thickness of the object to be measured is obtained.Furthermore, the relation function between the projection data and crosstalk ratio isobtained. Then the projection data can be corrected by the relation function. Correctionexperiments on the actual CT images show that this method can reduce the edge artifactsignificantly.

Related Dissertations

  1. Based on spectral regularization of linear dimensionality reduction methods,TP391.41
  2. The Research of Moving Object Detectionalgorithm in Intelligent Visual Surveillance System,TP391.41
  3. Excitation Method and FPGA Based Design and Implementation of Non-crosstalk Ultrasonic Ranging System,TP274.53
  4. Fuzzy C-Means Clustering and Rough Set coupled applications in slope stability analysis,TD824.7
  5. Investigation on Scatter Problem in Twin-axis Radiographic System,O436.2
  6. Research on Free Three-dimensional Display System Based on DMD Projection Technology and Cylindrical Lens,TP391.41
  7. The Conceptual Perspective comic humor language,H13
  8. Manifold-based Feature Extraction and Face Recognition Analysis,TP391.41
  9. Research of PS-InSAR Techology on Monitoring Surface Deformation in Xi’an,P227
  10. The Preliminary Study of Two Dimensional Ionization Chamber Array and Electronic Portal Imaging Device System in Dosimetric Verification,R730.5
  11. Edge technology based on double Doppler lidar system design,TN958.98
  12. The Algorithms Study of Beam-hardening Correction,TP391.41
  13. Research on X-ray On-line Quantitative Detection Technology for Welding Defect,TP274.4
  14. Research on Methods for Mechanical Failture Diagnosis Based on Blind Source Separation,TH17
  15. Research on the Signal Integrity of Carbon Nanotube Interconnects,TB383.1
  16. PMD and PDL Effects in Polarization Division Multiplexing Transmission,TN929.11
  17. CDEBP Neural Network and Researched on Its Application in Pre-assessments of the Automotive Wiring Harness Crosstalk,TP183
  18. Study and Development of Drying System for Waterish and Sticky Material,TQ051.8
  19. Electro-Thermal Characterization of Cabon Nanotube Interconnect and Its Impact on Propagating Signal,TB383.1
  20. Study of Town Sludge Stickiness in Drying Process,X703
  21. Theoretical and Numerical Study of Borehole Acoustic and Electric Field Excited by External Acoustic Sources under Seismoelectric Effects,P631.4

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