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