Dissertation > Excellent graduate degree dissertation topics show

Based on regularization super-resolution image sequence reconstruction

Author: ZhanMeiQuan
Tutor: DengZhiLiang
School: Jiangsu University of Science and Technology
Course: Applied Computer Technology
Keywords: Ill-posed problem Adaptive Regularization Super-resolution Image sequence reconstruction
CLC: TP391.41
Type: Master's thesis
Year: 2011
Downloads: 75
Quote: 0
Read: Download Dissertation

Abstract


In the current digital image application field, since the influence of physical conditions of the imaging system and climate conditions and other factors in the imaging process, there is often an optical blur and motion blur, the degradation process of the down-sampling and noise, etc., which makes the actual acquired image degradation the resolution of the image, resulting in up to less than the application requirements. The idea of ??super-resolution image sequence reconstruction techniques is the use of low-resolution image sequences to obtain high-resolution images, this is just to solve the current problems exist in the field of digital image application, super-resolution image sequence reconstruction function, however, has a morbid, in the actual application defects. Regularization method is an effective way to solve the ill-posed problem, regularization method for super-resolution image sequence reconstruction method in order to solve the above problems, a more comprehensive analysis and research, the main contents include: First, to design A new adaptive regularization super-resolution image sequence reconstruction methods. The traditional methods are assumed registration parameters estimation result is accurate, there is a certain degree of error in practical applications, however, the results of motion estimation. Maximum a posteriori probability estimates based on the super-resolution image reconstruction model, give full consideration to the registration parameters estimated error of reconstruction, choose the regularization parameter design integrated multiple model advantages of regular use of adaptive technology the objective function, regularization super-resolution image sequence reconstruction model was improved. Simulation results show that the new method makes reconstruction has a better image in the overall visual effect. Second, improved the L2 norm total variation regularization super-resolution image sequence reconstruction algorithm. Traditional regularization of super-resolution image sequence reconstruction algorithms assume that the relative contribution of each piece of low-resolution image reconstruction are equal, this will affect the reconstruction results. This article give full consideration to the relative contribution of each piece of the low resolution of the reconstruction and total variational regularization method used in the morbidly regularization to overcome the reconstruction problem, the use of the total variation of the super-resolution image sequence reconstruction, effectively maintaining The edges of the image. Simulation results showed that the improved algorithm is not only to improve the holding ability of the edges of the image, but also to more effectively suppress the noise, making the image clearer after reconstruction. Third, design a fast total variation regularization super-resolution image sequence reconstruction algorithm. Total variational regularization data items in the objective function in the form of the L2 norm, L2 norm, however, is more complex mathematical form. This paper the use of L1 norm instead of L2 norm of the traditional total change points is then the objective function to improve simulation results show the improved algorithm reconstruction speed than traditional algorithms reconstruction speed to be faster, and the reconstruction results with traditional algorithms considerable, even better.

Related Dissertations

  1. Design and Research on Stabilization Loop of Gyro Stabilized Pod Control System,V241.5
  2. Robust Adaptive Control Algorithms Research for the System of Spacecraft Attitude Dynamics,V448.22
  3. Research on Spread Spectrum Code Acquisition of Double Threshold Based on Sliding Correlator,TN914.42
  4. Study of Synchronization and Adaptive Frequency in Frequency-Hopping System,TN914.41
  5. Research on the Super-resolution Technologies of Polarimetric SAR Images,TN957.52
  6. Superresolution of Hyperspectral Images Based on Spatial-Spectral Information Coordination,TN911.73
  7. The Technology Research on Channel Coding and Channel Estimation in MC-CDMA Systems,TN929.533
  8. Video Coding Research Based on Texture Characteristics,TP391.41
  9. Research on Removal Algorithm of Shadows in Image Segmentation,TP391.41
  10. Study on Nonlinear Variable Structure Guidance Law,TJ765
  11. Optimal Synchronization Control of a Class of Underactuated Lagrangian Systems,TP13
  12. Adaptive Control of Post-Boost Vehicle Using Multiple Models,TP273.2
  13. Application and Research of Dual Control in Vertical Landing Aircraft,TP273
  14. Design of Gyro Stabilized Pod Control System,TP273
  15. Study on Adaptive Fuzzy Control Algorithm and Its Implementation,TP273.4
  16. Image-Guided Adaptive Radiation Therapy of Intensity-Modulated Radiation for Cervical Cancer,R737.33
  17. Research on synchronization control of chaotic systems,O415.5
  18. Research of Anomaly Detection Algorithms of Hyperspectral Imagery Based on Kernel Method,TP751
  19. Active Power Filter and Its Application in Distribution Network,TN713.8
  20. Research on Adaptive Control of Recovery of UUV with Near Wall Constraint,TP273
  21. Research and Design on Indoor Localization System Based on ZigBee,TN929.5

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