位置：成果数据库 > 期刊 > 期刊详情页
一种基于超完备字典学习的图像去噪方法
  • ISSN号：0732-2112
  • 期刊名称：电子学报
  • 时间：0
  • 页码：1-12
  • 语言：中文
  • 分类：TN911.73[电子电信—通信与信息系统;电子电信—信息与通信工程]
  • 作者机构：[1]中山大学数学与计算科学学院,广东广州510275, [2]中山大学信息科学与技术学院,广东广州510275, [3]广东省信息安全技术重点实验室,广东广州510275
  • 相关基金：国家自然科学基金（No.60675016,No.60633030）;国家973重点基础研究发展规划（No.2006CB303104）;国家自然科学基金与广东省政府联合资助重点项目（No.U0835005）
  • 相关项目：基于原象学习理论的人脸图像对齐方法
作者： 赖剑煌|蔡泽民|
关键词： 稀疏表示, 基追踪, 匹配追踪, 字典学习, 二次规划, sparse representation, basis pursuit, matching pursuit, dictionary learning, quadratic program
中文摘要：

基于超完备字典的图像稀疏表示因其具有稀疏性、特征保持性、可分性等特点而被广泛应用于图像处理.本文提出一种超完备字典学习算法并应用于图像去噪.将字典学习等价于一个二次规划问题,并提出适合于大规模运算的投影梯度算法.学习所得字典能有效描述图像特征.基于此超完备学习字典,获得图像的稀疏表示,并恢复原始图像.实验结果表明,与小波类去噪方法相比,本文的学习算法能更好地去除图像噪声,保留图像细节信息,获得更高的PSNR值.

英文摘要：

Images＇ sparse representations over over-complete dictionaries have a wide application in image processing due to the properties of sparsity,integrity and separability.This paper proposes a dictionary learning algorithm which is applied to image de-noising.The dictionary learning problem is expressed as a box-constrained quadratic program and a fast projected gradient method is introduced to solve it.The learned dictionary describes the image content effectively.Experimental results show that：in comparison with the wavelet-based de-noising methods,our learning-based algorithm has better de-noising ability,keep more detail image information and improve the peak signal-to-noise ratio.

关于赖剑煌：

人像识别技术的关键问题研究
期刊论文 13 会议论文 15
非线性聚类新方法及其在图像分析上的应用
期刊论文 40 会议论文 25
人体生物特征识别进展国际研讨会
基于原象学习理论的人脸图像对齐方法
期刊论文 13 会议论文 11 专利 1
面向视频监控的生物特征识别关键理论与技术研究
期刊论文 51 会议论文 46
非均匀光照下人脸稳定特征点的精确检测
期刊论文 19 会议论文 6
多摄像机复杂监控环境下的行人再标识关键问题研究
期刊论文 2
同期刊论文项目
数字媒体内容安全关键技术及评测方法的研究
期刊论文 81 会议论文 45 获奖 3 专利 42 著作 1
基于原象学习理论的人脸图像对齐方法
期刊论文 13 会议论文 11 专利 1
同项目期刊论文
基于二次多项式模型的人脸光照归一化
Penalized Pre-image Learning in Kernel Principal Component Analysis
Extraction of Illumination Invariant Facial Features from A Single Image Using Nonsubsampled Contour
生物特征数据安全保护技术的发展
1D-LDA vs. 2D-LDA: When is vector-based linear discriminant analysis better than matrix-based?
Image Segmentation and Selective Smoothing Based on Variational Framework
Perturbation LDA: Learning the difference between the class empirical mean and its expectation
Choosing parameters of kernel subspace LDA for recognition of face images under pose and illuminatio
一种基于最优局部信息融合的蛋白质亚细胞定位预测方法
分子系统发育分析的生物信息学方法
蛋白质亚细胞定位预测的机器学习方法
高等职业教育与职业资格证书制度融合的途径
Detecting Double JPEG Compression with the Same Quantization Matrix
Extraction of Illumination Invariant Facial Features from A Single Image Using Non-subsampled Contou
Efficient General Print-Scanning Resilient Data Hiding Based on Uniform Log-polar Mapping
基于非均匀DCT的量化索引调制隐写
Tensor-based Projection Using Ridge Regression and Its Application to Action Classification
Simultaneously estimating the fundamental matrix and homographies
An Experimental Study on the Security Performance of YASS.
Robust AVS Audio Watermarking.
Cryptanalysis and improvement of an ID-based ad-hoc anonymous identification scheme at CT-RSA 05
双线性对有效计算研究进展
JPEG Error Analysis and Its Applications to Digital Image Forensics.
Steganalysis of YASS
Edge Adaptive Image Steganography Based on LSB Matching Revisited.
Steganalysis of JPEG-CES
基于二次多项式模型的人脸光照归一化
A More Secure Steganography Based on Adaptive Pixel-Value Differencing Scheme
Fragility Analysis of Adaptive Quantization Based Image Hashing.
Semi-Fragile Zernike Moment-Based Image Watermarking for Authentication
Audio Watermarking Scheme Robust against Desynchronization Based on the Dyadic Wavelet Transform
Bypassing the decomposition attacks on two-round multivariate schemes by a practical cubic round
Why not use the LM method for fundamental matrix estimation?
计算双线性Weil对平方的一个注记
Normalization of Face Illumination Based on Large- and Small- Scale Features
Improving Robustness of Quantization-Based Image Watermarking via Adaptive Receiver
When Is Vector-based Linear Discriminant Analysis Better than Matrix-based?
All Pairings are in a Group
Audio Watermarking Robust to Time-Scale Modification and MP3 Compression
On the Forgeability of Wang-Tang-Li's ID-Based Restrictive Partially Blind Signature
Computing the Ate Pairing on Elliptic Curves with Embedding Degree k = 9
Invariant image watermarking based on statistical features in the low-frequency domain
A note on the Ate pairing
Steganalysis of multiple-base notational system steganography
Audio watermarking robust against time-scale modification and MP3 compression
Efficient Tate pairing computation using double-base chains
结合Zernike矩和水印的半脆弱图像认证.
Perturbation LDA: Learning Difference between the Class Empirical Mean and Its Expectation
A Study on the Randomness Measure of Image
Discriminating Computer Graphics Images and Natural Images Using Fractal Geometry.
不规则网格上的曲面设计方法
一种基于DDTW的三维碎片自动拼接方法
A Fast Two-Dimension 4×4 Inverse Integer Transform Algorithm for H.264/AVC
Forgeability of Wang-Tang-Li's ID-based restrictive partially blind signature scheme
基于标记隐藏的多媒体消息版权和内容监管
基于Gabor小波特征确认多媒体消息数字水印版权标识
一种保护芯片设计的多变量加密及其电路结构
使用阈下信道的可逆R-S数字水印
基于非均匀离散余弦变换的MPEG-4视频内容保护
Marching Cubes算法研究现状
结合Zernike矩和水印的半脆弱图像认证
基于内容自适应的优化DWT-HMM稳健 图像水印算法
Robust image watermarking based on multiband wavelets and empirical mode decomposition
Robust Image Watermarking Based on Multiband Wavelets and Empirical Mode Decomposition.
A Survey of Passive Technology for Digital Image Forensics
鲁棒的区域复制图像篡改检测技术.
信息安全综述
Survey of information security
鲁棒的区域复制图像篡改检测技术
用于视频多播传输的压缩传感实现方法研究
基于特征函数和质量因子的JPEG图像隐写分析
一种基于最优局部信息融合的蛋白质亚细胞定位预测方法
分子系统发育分析的生物信息学方法
蛋白质亚细胞定位预测的机器学习方法
基于非对称数据结构实现信息隐藏通信的方法研究