2017.08-Osher_LOW DIMENSIONAL MANIFOLD MODEL for IMAGE PROCESSING
2017-08-14 22:46
435 查看
摘要
本文提出了一个新的低维流形模型(low dimensional manifold model,简称LDMM)并将其用于图像处理问题中;LDMM基于这样的fact:很多自然图像的块流形有低维的结构;因此,可以用块流形的维数作为正则化来恢复图像。
LDMM中的关键是在point cloud上解决laplace-beltrami方程,可通过点积分(point integral)方法解决;
point interal方法正确的采用采样点的方法,结果比标准的graph laplacian好。
在图像去噪、修复、超分辨重建中的数值仿真表明LDMM是一个powerful method.
Introduction
图像恢复问题是不适定的,因此我们需要加正则化(也即先验);TV,可以很好的恢复图像的卡通部分,纹理部分却恢复的不好;
nonlocal TV,可以很好的恢复纹理;
本文motivation:图像 f 的patch set P(f) 是欧式空间的一个点集,P(f)采样自一个低维的光滑流形M(f),成为patch manifold。本文想去恢复原始的图像,使得图像的patch manifold M(f) 尽可能小,从而得到如下的优化模型
mindim(M(f))+λ||y−Φf||22
相关文章推荐
- Software and Hardware for Data Analysis, Pattern Recognition and Image Processing
- 数字图像处理-图像增强: MSRCR Method For Image Processing
- Visual Enhancement using Constrained L0 Gradient Image Decomposition for Low Backlight Displays
- Non-Photorealistic Rendering (Domain transform for edge-aware image and video processing)
- [转] Implementation of Fast Fourier Transform for Image Processing in DirectX 10
- Fast statistical calculations of sub matrices for image processing
- Gabor filter for image processing and computer vision
- Retinex processing for automatic image enhancement 翻译
- Image Processing for Dummies with C# and GDI+ Part 4 - Bilinear Filters and Resizing
- Image Processing Wavefronts for HEVC Parallelism
- 1MULTI-DIMENSIONAL SIGNAL PROCESSING AND CIRCUITS FOR ADVANCED ELECTRONICALLY SCANNED ANTENNA ARRAYS
- Charged Fluid Model for Brain Image Segmentation
- Using FPGAs for DSP Image Processing
- Image Processing for Dummies with C# and GDI+ Part 6 - The HSL color space
- Fundamentals of Three-Dimensional Digital Image Processing
- 4MULTI-DIMENSIONAL SIGNAL PROCESSING AND CIRCUITS FOR ADVANCED ELECTRONICALLY SCANNED ANTENNA ARRAYS
- C#图像处理教程 C# Tutorials for image processing
- Teaching: Excelent Website for Signal and image processing
- [论文笔记] Learning to Read Chest X-Rays Recurrent Neural Cascade Model for Automated Image Annotation
- Domain Transform for Edge-Aware Image and Video Processing - 论文阅读