Dissertation > Excellent graduate degree dissertation topics show

The Degaded Image Retoration Technology Research Based on Non-local Means

Author: ZuoNaNa
Tutor: HuZhengPing
School: Yanshan University
Course: Communication and Information System
Keywords: Image Restoration Single Image Atmospheric veil Super-resolution reconstruction Isomorphic to a dictionary to learn Non - local means Sparse Representation
CLC: TP391.41
Type: Master's thesis
Year: 2011
Downloads: 154
Quote: 0
Read: Download Dissertation

Abstract


How to restore the degraded image , to make it clear and hot research topics in the field of image processing, computer vision and pattern recognition . Recovery technology research for the field of astronomy , remote sensing, military surveillance , traffic monitoring , police security , medical diagnostic systems and other services , has important practical significance . Recovery of degraded image contains denoising , deblurring , restoration and super-resolution content , this article focused on a single image to fog with a single image super-resolution reconstruction of two aspects of conducting research . First, an overview of the general principles of image restoration techniques , and build the the two degraded image degradation model : fog degraded image degradation model with low-resolution degraded image degradation model ; Next, we discuss the non- local means filtering algorithm . Secondly, the the demister algorithm for the current lead to the edge of the halo effect , edge contour and scenery features more obscure given scene depth image restoration algorithm based on non - local mean filtering single fog-degraded prior knowledge under unknown conditions . First, with the non - local filtering mist average pretreatment estimated sky brightness ; Secondly, according to the high frequency of the edge with the fog image atmospheric veil having a large similarity Nonlocal filtering algorithm to estimate the atmospheric veil , avoiding difficult to find the scene depth ; Fig Finally, prevent the smooth and chrominance contrast amplification adjustment processing . Experiments show that , given algorithm can not only make the recovery image edge contour and landscape characteristics are relatively clear , and can effectively suppress the edge of the halo effect . Finally, most of the low signal-to-noise ratio super-resolution reconstruction algorithm result is not satisfactory , and the majority of the algorithm based on the reconstruction of the gray image , given a single image super-resolution based on non - local means filtering and sparse The reconstruction algorithm . First, using the image-based block sparse representation method, training isomorphic ultra complete dictionary ; the luminance domain sparse Secondly, through the coefficient of sparse representation Reconstruction and their Nonlocal filtering process ; Finally, UV chrominance Reconstruction . The experiments show that the given algorithm effectively suppressed edge halo , artificial artifacts , robustness to noise .

Related Dissertations

  1. Hypercomplex sparse representation and its applications,TP391.41
  2. Research of Digital Image Inpainting Technology,TP391.41
  3. Synthetic aperture imaging quality optical system recovery method,TP391.41
  4. Based on Genetic Algorithm super- resolution image reconstruction study,TP391.41
  5. Sparse Representation Fordimensionality Reduction,TP391.41
  6. Confocal Microscopy Image Restoration and Regularization Methods,TP391.41
  7. Deconvolutional Networks for Image Representation and Restoration,TP391.41
  8. Research on Sparse Classification Method of Facial Image,TP391.41
  9. Sparse representation and discriminant analysis techniques applied research in face recognition,TP391.41
  10. Compression of Transmission Line Insulator Leakage Current Data Based on Compressive Sensing Theory,TM75
  11. Research and Application of the Panoramic Annular Lens Imaging System,TP391.41
  12. Based on sparse representation of off-line handwritten Chinese character recognition,TP391.1
  13. Common types blurred image restoration Research and Implementation,TP391.41
  14. Research of Underwater Image Restoration Algorithm Based on Backward Scattering Noise Model,TP391.41
  15. Research on Algorithm of Image Super-resolution Reconstruction,TP391.41
  16. Research on Digital Image Inpainting,TP391.41
  17. Remote sensing image reconstruction algorithm oriented IICCD camera is not completely random sampling,TP751
  18. Image restoration of key technologies,TP391.41
  19. The Research of Image Restoration Based on Phase Diversity Method,TP391.41
  20. Based on non-local means filtering denoising ultrasound images and image sequences,TP391.41
  21. Research on Signal Recoveryalgorithms Based on Compressed Sensing,TN911.7

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