应用通用自回归模型实现图像的自适应滤波

郝飞; 史金飞; 张志胜; 陈茹雯

doi:10.3788/OPE.20142201.0186

您当前的位置：

首页 >

文章列表页 >

应用通用自回归模型实现图像的自适应滤波

信息科学 | 更新时间：2020-08-13

- 应用通用自回归模型实现图像的自适应滤波
- Image adaptive filtering using general auto-regressive model
- 光学精密工程 2014年22卷第1期页码：186-192
- 作者机构：
  
  1. 东南大学机械工程学院,江苏南京,211189
  2. 南京工程学院机械工程学院,江苏南京,211167
- 作者简介：
- 基金信息：
  
  国家自然科学青年基金资助项目(No. 51205182)();江苏省高校自然科学基础研究项目(No.12KJB510006)();南京工程学院创新基金面上项目(No.C
- DOI：10.3788/OPE.20142201.0186
  中图分类号： TP391.4
- 收稿日期：2013-05-10，
  
  纸质出版日期：2014-01-15
- 稿件说明：
移动端阅览
郝飞, 史金飞, 张志胜等. 应用通用自回归模型实现图像的自适应滤波[J]. 光学精密工程, 2014,22(1): 186-192

HAO Fei, SHI Jin-fei, ZHANG Zhi-sheng etc. Image adaptive filtering using general auto-regressive model[J]. Editorial Office of Optics and Precision Engineering, 2014,22(1): 186-192
郝飞, 史金飞, 张志胜等. 应用通用自回归模型实现图像的自适应滤波[J]. 光学精密工程, 2014,22(1): 186-192 DOI： 10.3788/OPE.20142201.0186.

HAO Fei, SHI Jin-fei, ZHANG Zhi-sheng etc. Image adaptive filtering using general auto-regressive model[J]. Editorial Office of Optics and Precision Engineering, 2014,22(1): 186-192 DOI： 10.3788/OPE.20142201.0186.

摘要

考虑数字图像滤波处理对融线性和非线性于一体的数学模型的需求

根据Weierstrass逼近理论推导建立了通用的自回归数学模型。该模型将线性自回归模型和非线性自回归模型融合于一个统一的数学表达式中

仿真实验表明其能够较好地拟合现有的线性和非线性自回归模型。用二维向量取代标量参数

推导了通用自回归模型的二维数学表达式。通过对比分析

确定采用GM(Generalized M estimator)参数估计法进行参数估计。实验结果表明

该算法收敛较快

平均迭代次数不超过6次

线性模型平均计算耗时为150 s

二次模型平均耗时为418 s。提出的二维通用自回归模型滤波方法能较好地保留图像的细节信息

图像滤波效果好。

Abstract

As the model fused a linear model and a nonlinear model is beneficial to digital image filtering

this paper explores a generalized autoregressive model on the basis of Weierstrass theory for image adaptive filtering. The model fuses both linear and nonlinear autoregressive models into a uniform expression and simulation experiments verify that the model can fit both conventional linear and nonlinear autoregressive models well. By using a bi-vector instead of a scalar parameter

the bi-dimensional expression of the model is deduced

then a generalized M-estimator is chosen to estimate parameters by a contrast analysis. The experimental results indicate that the proposed algorithm has a fast convergence speed

the average iterations are no more than 6 times and the computing time for linear model and quadratic model is 150 s and 418 s respectively. Moreover

it can remove image noises while conserve detailed image information effectively.

关键词

Keywords

references

FAN H J, SONG Q. A sparse kernel algorithm for online time series data prediction [J]. Expert Systems with Applications, 2013, 40(6): 2174-2181.

STEPNICKA M, CORTEZ P, DONATE J P, et al.. Forecasting seasonal time series with computational intelligence: On recent methods and the potential of their combinations [J]. Expert Systems with Applications, 2013, 40(6): 1981-1992.

TANG R, SHAO J, ZHANG Z J. Sparse moving maxima models for tail dependence in multivariate financial time series [J]. Journal of Statistical Planning and Inference, 2013, 143(5): 882-895.

ZACCARELLI N, LI B L, PETROSILLO I, et al.. Order and disorder in ecological time-series: Introducing normalized spectral entropy [J]. Ecological Indicators, 2013, 28(SI): 22-30.

刘志兵, 王西彬. 积屑瘤状态对微细切削表面轮廓特征的影响[J]. 光学精密工程, 2011, 19(1): 90-96.

LIU ZH B, WANG X B. Influence of built-up edge phases on characteristics of surface profile of micro cutting [J]. Opt. Precision Eng., 2011, 19(1): 90-96. (in Chinese)

TONG H, LIM K S. Threshold autoregression, limit-cycles and cyclical data [J]. Journal of the Royal Statistical Society Series B-Methodological, 1980, 42(3): 245-292.

HAGGAN V, OZAKI T. Amplitude-Dependent Exponential AR Model Fitting for Non-linear Random Vibrations [M]. Time Series. Amsterdam: North-Holland. 1980: 57-71.

GRANGER C, ANDERSON A. An introduction to bilinear time-series models [J]. International Statistical Review, 1980, 48(2): 238.

KASHYAP R L, EOM K B. Robust images techniques with an image restoration application [J]. IEEE Transactions on Acoustics, Speech and Signal Processing, 1988, 36(8): 1313-1325.

ALLENDE H, GALBIATI J, VALLEJOS R. Robust image modeling on image processing [J]. Pattern Recognition Letters, 2001, 22(11): 1219-1231.

AMIRMAZAGHANI M, AMINDAVAR H. A novel statistical approach for speckle filtering of SAR images [C].IEEE 13th Digital Signal Processing Workshop & 5th IEEE Processing Education Workshop, Marco Isl, FL, JAN 04-07, 2009: 457-462.

BUSTOS O, OJEDA S, VALLEJOS R. Spatial ARMA models and its applications to image filtering [J]. Brazilian Journal of Probability and Statistics, 2009, 23(2): 141-165.

TAKALO R, HYTTI H, IHALAINEN H. Adaptive autoregressive model for reduction of poisson noise in scintigraphic images [J]. Journal of Nuclear Medicine Technology, 2011, 39(1): 19-26.

施招云, 万德钧, 黄仁. 一种非线性时序模型的结构辨识方法 [J]. 合肥工业大学学报:自然科学版, 1993, 16(S1): 44-49.

SHI ZH Y, WAN D J, HUANG R. A structure identification method of nonlinear time series models [J]. Journal of Hefei University of Technology:Natural Science, 1993, 16(S1): 44-49. (in Chinese)

HA E, NEWTON H J. The bias of estimators of causal spatial autoregressive processes [J]. Biometrika, 1993, 80(1): 242-245.

魏彤, 郭蕊. 自适应卡尔曼滤波在无刷直流电机系统辨识中的应用[J]. 光学精密工程, 2012, 20(10): 2308-2314.

WEI T, GUO R. Application of adaptive identification of Kalman filtering to system brushless DC motor [J]. Opt. Precision Eng., 2012, 20(10): 2308-2314. (in Chinese)

冯肖维, 何永义, 方明伦, 等. 应用特征估计的距离图像多尺度滤波[J]. 光学精密工程, 2011, 19(5): 1118-1125.

FENG X W, HE Y Y, FANG M L, et al.. Multi-scale smoothing of noisy ranges image using feature estimation[J]. Opt. Precision Eng., 2011, 19(5): 1118-1125. (in Chinese)

浏览量

387

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

基于双目结构光周视扫描的3D件三维重构与位姿估计研究

重叠分数与匹配矩阵驱动的部分重叠点云配准网络设计

基于Snell窗及偏振模式的水下载体航姿测量

改进多尺度结构化融合的红外与可见光图像融合

非成对训练样本条件下的红外图像生成