图像采集系统及其小波域快速PCT去噪算法
|
摘要
1. 引言
图像采集和预处理是图像处理领域最重要技术之一。图像采集的目的是把现实世界的模拟图像采到处理通道中并转化成数字图像，这一过程的关键在于如何快速实时地把模拟图像转化成数字图像；而预处理的目的是通过对图像的去噪处理来提高信噪比，突出图像的期望特征。图像去噪方法有时域和频域两类方法，但归根到底都是利用噪声和信号在频域上的分布不同进行去噪的。信号主要分布在低频区域，而噪声主要分布在高频区域，同时图像的细节也分布在高频区域。所以，图像信号去噪中存在一个一直尚未得到很好解决的困难，即在去除图像随机噪声的同时引进图像边缘的模糊，在保留和增强图像边缘的同时又增大了图像的噪声。传统的低通滤波方法将图像的高频部分滤除，虽然能达到降低噪声的效果，但会破坏图像的细节。如何构造一种既能够降低图像噪声，又能够保持图像细节的去噪方法，一直是图像处理领域中理论和实践工作者的研究目标。小波变换这一强有力的信号分析工具的出现，使上述目标的实现成为可能。
小波理论是近年来新兴的一门应用理论，并得到了非常迅速的发展，而且由于其具备良好的时频特性，因而实际应用也非常广泛。传统的傅里叶变换有明显的缺陷——无时间局部信息。而实际信号往往是时变信号、非平稳过程，了解它们的局部特性是很重要的。小波本身的特性使它特别适用于分析非平稳信号。小波分析的应用研究是与小波分析的理论研究紧密地结合在一起的。同时，在小波领域又能够划分出很多的分支领域，多进制小波、小波包基、分解尺度、抽取方法等等。可见，在小波不仅功能强大，还有很多亟待解决的问题在等着我们去发现去研究。
2. 图像采集系统硬件设计
作者首先考察了当前图像采集系统的系统结构，分析了各种系统结构的优缺点，在此基础上作者设计了一种快速、实时的图像采集系统。本系统采用了PHILIPS公司地高速视频解码芯片SAA7111进行A/D转换，以Altera公司的大规模集成FPGA芯片EP1K10TC144作为图像采集控制模块，数据处理芯片
选用了TI公司的TMS320C6204。同时，系统中设置了两片大容量的SRAM作为帧存储器，并采用乒乓读写机制以提高存储器的读写速度。
3. 去噪算法
本文工作的重点集中在图像去噪算法的研究，介绍了当前信号处理领域尤其是图像处理领域去噪算法的发展状况，通过探讨时域和频域的图像去噪方法，进而确定了本文在小波域进行信号处理的基本方针。在众多小波去噪算法中，阈值萎缩法是最常用方法，利用阈值法进行信号去噪处理很大程度上取决于阈值门限和阈值函数的选取。所以又考察了小波去噪算法中对阈值门限和阈值函数的选取理论。阈值法尽管对有些图像的去噪效果好，但从普遍意义上，去噪效果要差一些，在应用上有一定的局限性。因此，开始寻找去噪的新方法，通过对比例萎缩法去噪原理和去噪实例的分析，发现比例法恰好能弥补阈值法的缺点，作者在此基础上提出一种新的小波去噪算法——PCT法，实验证明其适应性和效果上要远远优于阈值法及其改进方法。
作者从以上几个不同的角度入手，确立了本文基本的图像去噪研究思想：噪声-小波-阈值法-比例萎缩法-PCT法去噪。针对小波变换的紧支性和去相关性等特性，去噪算法将从噪声干扰中恢复真实的信号。其去噪主要步骤如下：
第一步，对图像信号进行三级离散小波变换，得到小波变换系数，其中，k表示分解层数；i，j表示象素点位置。
第二步，估计信号小波系数的方差，其计算式为
第三步，利用第二步得到的来计算PCT法中的阈值门限，其计算式为。
第四步，运用PCT法的函数式，求出处理后的小波系数，其函数式表示为
第五步，对进行离散小波反变换，从而得到去噪后的信号。
4. 结论
通过以上几个方面的研究工作，重点提出了小波变换快速算法和PCT去噪算法。在计算机仿真实验中，对Woman图像加噪信号进行处理，并且同常用的硬阈值、软阈值法，硬、软阈值折衷法，导数连续的阈值法及比例萎缩法的结果比较，结果比较，通过对去噪前后的峰值信噪比(PSNR)和均方误差(MSE)的变化情况说明这种去噪算法对于图像噪声的去除是有效的，尤其是在低信噪比时效果更为明显。
1. Introduction
The gathering and the pretreatment of image is one of the most important technology of image process fields. The purpose that the image is gathered is to transforms the analogical image of the real world into a digital image, the key to this course lies in how to transform the analogical image into a digital image fast. And the purpose of the pretreatment is to improve SNR to stress the image characteristic through denoising processing. Image denoising includes space (time) and frequency domain methods. But their essence is to use the differences between noise and signal distributing in frequency domain, the signal is mainly distributing in low frequency part, while noise and image detail mainly in high frequency part. So there always is a difficulty needs to be solved, it’s to eliminate random noise but not causing image edge blurring, to keep and increase image edge but not enlarge noise. The traditional low-pass filter method can restrain the high frequency part and decrease noise but may destroy image detail, and other methods also have other defects. How to construct a denoising algorithm which decreasing noise and keep image detail is always research object of scientists in image processing world. After the appearance of new signal processing tool—wavelet, the object can be implemented easily in reality.
Wavelet analysis is a rapid developed new area of nowadays applied mathematics. Wavelet transform(WT) is space(time) and frequency local transform compared to Fourier transform and Windowed Fourier transform(Gabor transform). So it can pick-up local information from signal and can realize multi-analysis to signal by stretching 、 shrinking and shift etc. It solved many problems which can’t be solved by Fourier transform. So we call wavelet transform as “mathematics microscope”. Now wavelet transform is used widely in many areas. Especially in image processing area it shows great advantage and has promising future.
2. The Hardware Design of Image Gathering System
In this paper, the author studied the structure of the picture gathering system at first, analysed the merit and demerit of different system. On this foundation, the author designed a
real-timely picture gathering system. This system adopts the PHILIPS Company's high-speed video decode chip SAA7111 to perform A/D transition, used the integrated FPGA chip EP1K10TC144 of Altera Company as controlling the picture gathering, and the TMS320C6204 of TI Company was selected for data processing chip. Meanwhile, two SRAM of large capacity were used as the frame memories, and author adopted the ping-pong reading and writing mechanism to improving the speed of reading and writing memories.
3. Denoising Algorithm
This emphasis of this paper's job concentrate on the research of image denoising algorithm. Author recommend the state of development of the denoising algorithm at present in signal processing field, especially image processing field. After the research of denoising algorithms in space(time) and frequency domain, the paper proposed a denoising algorithm in wavelet domain. Of all wavelet denoising algorithms, threshold-method is used the most usually. Threshold-method depends on the selection of threshold value and threshold function at great extent. The author studied the theories of selection of threshold value and threshold function in the course. Though threshold method can denoise effectually to some images, from universal significance, it can not do well, and have some limitation on using. So author looked for new denoising method, and found that proportion-shrinking method can remedy the demerit of threshold method through analysis to principle and examples of proportion-shrinking denoising algorithm. On it a new wavelet denoising algorithm--PCT was proposed by author, the experiment conclusion proves its adaptability and denoising effect far superior to threshold method and improved threshold method.
The author established the base denoising research thought from different angles discussed above: noise-wavelet
图像采集和预处理是图像处理领域最重要技术之一。图像采集的目的是把现实世界的模拟图像采到处理通道中并转化成数字图像，这一过程的关键在于如何快速实时地把模拟图像转化成数字图像；而预处理的目的是通过对图像的去噪处理来提高信噪比，突出图像的期望特征。图像去噪方法有时域和频域两类方法，但归根到底都是利用噪声和信号在频域上的分布不同进行去噪的。信号主要分布在低频区域，而噪声主要分布在高频区域，同时图像的细节也分布在高频区域。所以，图像信号去噪中存在一个一直尚未得到很好解决的困难，即在去除图像随机噪声的同时引进图像边缘的模糊，在保留和增强图像边缘的同时又增大了图像的噪声。传统的低通滤波方法将图像的高频部分滤除，虽然能达到降低噪声的效果，但会破坏图像的细节。如何构造一种既能够降低图像噪声，又能够保持图像细节的去噪方法，一直是图像处理领域中理论和实践工作者的研究目标。小波变换这一强有力的信号分析工具的出现，使上述目标的实现成为可能。
小波理论是近年来新兴的一门应用理论，并得到了非常迅速的发展，而且由于其具备良好的时频特性，因而实际应用也非常广泛。传统的傅里叶变换有明显的缺陷——无时间局部信息。而实际信号往往是时变信号、非平稳过程，了解它们的局部特性是很重要的。小波本身的特性使它特别适用于分析非平稳信号。小波分析的应用研究是与小波分析的理论研究紧密地结合在一起的。同时，在小波领域又能够划分出很多的分支领域，多进制小波、小波包基、分解尺度、抽取方法等等。可见，在小波不仅功能强大，还有很多亟待解决的问题在等着我们去发现去研究。
2. 图像采集系统硬件设计
作者首先考察了当前图像采集系统的系统结构，分析了各种系统结构的优缺点，在此基础上作者设计了一种快速、实时的图像采集系统。本系统采用了PHILIPS公司地高速视频解码芯片SAA7111进行A/D转换，以Altera公司的大规模集成FPGA芯片EP1K10TC144作为图像采集控制模块，数据处理芯片
选用了TI公司的TMS320C6204。同时，系统中设置了两片大容量的SRAM作为帧存储器，并采用乒乓读写机制以提高存储器的读写速度。
3. 去噪算法
本文工作的重点集中在图像去噪算法的研究，介绍了当前信号处理领域尤其是图像处理领域去噪算法的发展状况，通过探讨时域和频域的图像去噪方法，进而确定了本文在小波域进行信号处理的基本方针。在众多小波去噪算法中，阈值萎缩法是最常用方法，利用阈值法进行信号去噪处理很大程度上取决于阈值门限和阈值函数的选取。所以又考察了小波去噪算法中对阈值门限和阈值函数的选取理论。阈值法尽管对有些图像的去噪效果好，但从普遍意义上，去噪效果要差一些，在应用上有一定的局限性。因此，开始寻找去噪的新方法，通过对比例萎缩法去噪原理和去噪实例的分析，发现比例法恰好能弥补阈值法的缺点，作者在此基础上提出一种新的小波去噪算法——PCT法，实验证明其适应性和效果上要远远优于阈值法及其改进方法。
作者从以上几个不同的角度入手，确立了本文基本的图像去噪研究思想：噪声-小波-阈值法-比例萎缩法-PCT法去噪。针对小波变换的紧支性和去相关性等特性，去噪算法将从噪声干扰中恢复真实的信号。其去噪主要步骤如下：
第一步，对图像信号进行三级离散小波变换，得到小波变换系数，其中，k表示分解层数；i，j表示象素点位置。
第二步，估计信号小波系数的方差，其计算式为
第三步，利用第二步得到的来计算PCT法中的阈值门限，其计算式为。
第四步，运用PCT法的函数式，求出处理后的小波系数，其函数式表示为
第五步，对进行离散小波反变换，从而得到去噪后的信号。
4. 结论
通过以上几个方面的研究工作，重点提出了小波变换快速算法和PCT去噪算法。在计算机仿真实验中，对Woman图像加噪信号进行处理，并且同常用的硬阈值、软阈值法，硬、软阈值折衷法，导数连续的阈值法及比例萎缩法的结果比较，结果比较，通过对去噪前后的峰值信噪比(PSNR)和均方误差(MSE)的变化情况说明这种去噪算法对于图像噪声的去除是有效的，尤其是在低信噪比时效果更为明显。
1. Introduction
The gathering and the pretreatment of image is one of the most important technology of image process fields. The purpose that the image is gathered is to transforms the analogical image of the real world into a digital image, the key to this course lies in how to transform the analogical image into a digital image fast. And the purpose of the pretreatment is to improve SNR to stress the image characteristic through denoising processing. Image denoising includes space (time) and frequency domain methods. But their essence is to use the differences between noise and signal distributing in frequency domain, the signal is mainly distributing in low frequency part, while noise and image detail mainly in high frequency part. So there always is a difficulty needs to be solved, it’s to eliminate random noise but not causing image edge blurring, to keep and increase image edge but not enlarge noise. The traditional low-pass filter method can restrain the high frequency part and decrease noise but may destroy image detail, and other methods also have other defects. How to construct a denoising algorithm which decreasing noise and keep image detail is always research object of scientists in image processing world. After the appearance of new signal processing tool—wavelet, the object can be implemented easily in reality.
Wavelet analysis is a rapid developed new area of nowadays applied mathematics. Wavelet transform(WT) is space(time) and frequency local transform compared to Fourier transform and Windowed Fourier transform(Gabor transform). So it can pick-up local information from signal and can realize multi-analysis to signal by stretching 、 shrinking and shift etc. It solved many problems which can’t be solved by Fourier transform. So we call wavelet transform as “mathematics microscope”. Now wavelet transform is used widely in many areas. Especially in image processing area it shows great advantage and has promising future.
2. The Hardware Design of Image Gathering System
In this paper, the author studied the structure of the picture gathering system at first, analysed the merit and demerit of different system. On this foundation, the author designed a
real-timely picture gathering system. This system adopts the PHILIPS Company's high-speed video decode chip SAA7111 to perform A/D transition, used the integrated FPGA chip EP1K10TC144 of Altera Company as controlling the picture gathering, and the TMS320C6204 of TI Company was selected for data processing chip. Meanwhile, two SRAM of large capacity were used as the frame memories, and author adopted the ping-pong reading and writing mechanism to improving the speed of reading and writing memories.
3. Denoising Algorithm
This emphasis of this paper's job concentrate on the research of image denoising algorithm. Author recommend the state of development of the denoising algorithm at present in signal processing field, especially image processing field. After the research of denoising algorithms in space(time) and frequency domain, the paper proposed a denoising algorithm in wavelet domain. Of all wavelet denoising algorithms, threshold-method is used the most usually. Threshold-method depends on the selection of threshold value and threshold function at great extent. The author studied the theories of selection of threshold value and threshold function in the course. Though threshold method can denoise effectually to some images, from universal significance, it can not do well, and have some limitation on using. So author looked for new denoising method, and found that proportion-shrinking method can remedy the demerit of threshold method through analysis to principle and examples of proportion-shrinking denoising algorithm. On it a new wavelet denoising algorithm--PCT was proposed by author, the experiment conclusion proves its adaptability and denoising effect far superior to threshold method and improved threshold method.
The author established the base denoising research thought from different angles discussed above: noise-wavelet
引文
[1]林鹏，图像采集卡的发展状况，中国图象图形学报，1996，1(2)：166-169
[2]田宏强，基于PCI总线的图像采集卡的开发与应用，微计算机信息，2000，16(3)：28-30
[3]陈冰等，基于TMS320C6x的实时图像处理系统，光电工程，2000，12，27(6)：37-42
[4]荆仁杰，叶秀清， 计算机图像处理.杭州： 浙江大学出版社，1990，6：1-34
[5]容观澳，计算机图像处理，北京：清华大学出版社，2000，2：1-69
[6]孙仲康，沈振康，Visual C++数字图像处理，北京：电子工业出版社，1999：527-561
[7]李叔梁等译，数字图像处理，北京：科学出版社，1982：1-54
[8]杨福生，小波变换的工程分析与应用，北京：科学出版社，1999：1-176
[9]程正兴，小波分析算法与应用，西安：西安交通大学出版社，1998：1-52
[10]赵松年，熊小芸，子波变换与子波分析，北京：电子工业出版社，1996：1-80
[11]崔锦泰，小波分析导论，西安：西安交通大学出版社，1995：1-24
[12]秦前清，杨宗凯，实用小波分析，西安：西安电子科技大学出版社，1994：1-16
[13]彭玉华，小波变换与工程应用，北京：科学出版社，1999：1-62
[14]李建平主编，小波分析与信号处理――理论、应用及软件实现，重庆：重庆出版社，1997：3-137
[15]陈逢时，子波变换理论及其在信号处理中的应用，北京：国防工业出出版社，1998：20-162
[16]Wang DCC, Vagnucci A H, LCC. Gradient inverse weighted smoothing scheme and evaluation of its performance. 1981, CGIP 16: 167-181
[17] Tukey J.W. Exploratory data analysis. Addison-Wesley, 1977
[18] Lin Yin, Ruikang Yang,Moncef Gabbouj, Yrj? Neuvo.Weighted Median Filters: A Tutorial. IEEE Transaction on circuits and systems-Ⅱ:Analog and digital signal processing,1996,43: 157-191
[19]Jaakko Astola, Petri Haavisto, Yrj? Neuvo. Vector median filter. Proceedings of the IEEE, 1990, 78: 678-689
[20]陈贺新，非线性滤波器与数字图像处理，北京：国防工业出版社，1997，9：1-23
[21]Tao Chen, Kai-Kuang Ma,Li-Hui Chen. Tri-state median filter for image denoising. IEEE Trans. image processing, 1999,8:1834-1838
[22]D. L.Donoho and I. M. Johnstone, "Wavelet shrinkage: Asymptopia", J. R. Stat. Soc. B, 1995, 57: 301-369
[23]D. L. Donoho and I. M. Johnstone, "Ideal time-frequency denoising," Technical Report, Dept. of Statistics, Standford University ,1994
[24]D. L. Donoho and I.M. Johnstone, "Ideal spatial adaptation via wavelet shrinkage," Biometrika,1994, 81: 425-455
[25]D. L. Donoho, "De-noising by soft thresholding," IEEE Trans. Information Theory, 1995, 41: 613-627
[26] PHILIPS Corporation. SAA7111 Data Sheet [Z]. 1998
[27]Altera Corporation. ACEX 1K Programmable Logic Device Family Data Sheet. 2003. Ver3.4
[28]刘静、常丹华，ACEX 1K系列CPLD的配置，重庆工商大学学报，2003，3，20(1)：44-46
[29]何伟等，FPGA/CPLD可编程逻辑器件的在系统配置方法，2003，5，26(5)：125-128
[30]刘德良等，多分辨率图像实时采集系统的FPGA逻辑设计，电子技术应用，2003，3：69-72
[31]Kassim A A, Chua K S, Fong F K, etal. DSP-based system for real-time video communications [J]. International Journal of Imaging Systems and Technology, 1999, 10(1): 47-53
[32]TI company. TMS320C62x/C67x Technical Brief, 1998
[33]TI company. TMS320C62x/C67x CPU and Instruction Set Reference Guide, 1998
[34]IDT Corporation. IDT71V424L/S Data Sheet. 2003
[35]AMCC S5933 PCI Controller Data Book, 1996
[36]曹明，陈文正，PCI总线协议的FPGA实现及驱动设计，电子技术应用，2000，7：15-18
[37]李贵山、戚德虎，PCI局部总线开发者指南[M]，西安：西安电子科技大学出版社，1997
[38]Kamel Belkacem-Boussaid, Azeddine Beghdadi, A new image smoothing
method basdon a simple model of spatial processing in the early stages of human vision. IEEE Trans. Image Processing, 2000, 9: 220-226
[39]M. K. Mihcak, I. Kozintsev, K.Ramchandran, “Spatially adaptive statistical modeling of wavelet image coefficients and its application to denoising”, in Proc. IEEE Int. Conf. Acoust, Speech, and Signal Proc, Phoenix, AZ, March 1999, 6: 3253-3256
[40]M. K. Mih?ak, I. Kozintsev, K. Ramchandran, P. Moulin. “Low-complexity image denoising based on statistical modeling of wavelet coefficients”. IEEE Trans. on Signal Processing, December 1999, 6(12): 300-303
[41]胡昌华、张军波、夏军等，基于MATLAB的系统分析与设计－小波分析，西安：西安电子科技大学出版社，1999，12：241-246
[42]王艳辉，王永臣，小波变换技术用于图像信号的降噪，沈阳工业学院学报，2000，12，19 (4)：2-15
[43]吴小培，黄端旭，汪炳汉，图像细节保持中值滤波器，图像图形学论文集－第五届全国图像图形学学术会议，1：38-43
[44]D. L. Donoho and I. M. Johnstone, “Adapting to unknown smoothness via wavelet shrinkage”, Journal of the American Statistical Associatio,1995, 90: 1200-1224
[45]李先涛，基于小波变换的维纳滤波器图像去噪方法的研究，吉林大学硕士学位论文，2003
[46]刘刚，屈梁生，自适应阈值选择和小波消噪方法的研究，信号处理，2002，12，18(6)：509-512
[47]曲天书等，基于SURE无偏估计的自适应小波阈值去噪，电子学报，2002，2，30(2)：266-268
[48]Xiao-ping Zhang, M. Desai. “Adaptive denoising based on SURE risk”, [J]. IEEE Signal Processing Letters. October 1998, 5: 265-267
[49]Xiao-ping Zhang, M. Desai, “Nonlinear adaptive noise suppression based on wavelet transform”, In: Proceeding of the IEEE International Conference on Acoustics, and Signal Processing[C], 1997: 1589-1592
[50]G. P. Nason, “Choice of the threshold parameter in wavelet function estimation”, Wavelets and Statistics, vol.103, LNS: 261-280, New York, 1995. Springer-Verlag
[51]Weyrich N, Warhola G T, “Wavelet shrinking and generalized cross validation for image denoising”, [J]. IEEE Trans. Image Processing, 1998, 7(1): 82-90
[52]Jansen M, Malfait M, Bultheel A, “Generalized cross validation for wavelet thresholding”, Signal Processing, 1997, 56(1): 33-44
[53]谢杰成等，小波图像去噪综述，中国图象图形学报，2002，3，7(3)：209-217
[54]谢杰成等，一种小波去噪方法的几点改进，清华大学学报，2002，42(9)：1269-1272
[55]Vidakovic B, Lozoya C B, “On time-depentent wavelet denoising”, [J]. IEEE Trans. Signal Processing, 1998, 46(9): 2549-2551
[56]Ching P C, So H C, Wu S Q, “On wavelet denoising and its application to time delay estimation”, [J]. IEEE Trans. Signal Processing, 1993, 47(10): 2879-2882
[57]Shark L K, Yu C, “Denoising by optimal fuzzy thresholding in wavelet domain”, [J]. Electronics Letters, 2000, 36(6): 581-582
[58]Malfait M, Roose D, “Wavelet-based image denoising using a Markov random field a priori mode”, [J]. IEEE Trans. Image Processing, 1997, 6(4): 549-565
[59]Mallat S, “A theory for multiresolution signal decomposition: the wavelet representation”,[J]. IEEE Trans. On Pattern Analysis and Machine Intelligence, 1989, 11(7): 674-693
[60]Sweldens W, “The Lifting Scheme: a Construction of Second Generation Wavelet”, [J]. SIAM J Math Anal, 1997, 29(2): 511-546
[61]周宁等，提升小波快速算法及其在JPEG2000中的应用，中国有线电视，2002，(18)：6-10
[62]程佩青，数字信号处理教程，北京：清华大学出版社，1998：215-252
[63]虞湘宾、毕光国，长序列信号快速相关及卷积的算法研究，电路与系统学报，2001，12，6(4)：78-8
[2]田宏强，基于PCI总线的图像采集卡的开发与应用，微计算机信息，2000，16(3)：28-30
[3]陈冰等，基于TMS320C6x的实时图像处理系统，光电工程，2000，12，27(6)：37-42
[4]荆仁杰，叶秀清， 计算机图像处理.杭州： 浙江大学出版社，1990，6：1-34
[5]容观澳，计算机图像处理，北京：清华大学出版社，2000，2：1-69
[6]孙仲康，沈振康，Visual C++数字图像处理，北京：电子工业出版社，1999：527-561
[7]李叔梁等译，数字图像处理，北京：科学出版社，1982：1-54
[8]杨福生，小波变换的工程分析与应用，北京：科学出版社，1999：1-176
[9]程正兴，小波分析算法与应用，西安：西安交通大学出版社，1998：1-52
[10]赵松年，熊小芸，子波变换与子波分析，北京：电子工业出版社，1996：1-80
[11]崔锦泰，小波分析导论，西安：西安交通大学出版社，1995：1-24
[12]秦前清，杨宗凯，实用小波分析，西安：西安电子科技大学出版社，1994：1-16
[13]彭玉华，小波变换与工程应用，北京：科学出版社，1999：1-62
[14]李建平主编，小波分析与信号处理――理论、应用及软件实现，重庆：重庆出版社，1997：3-137
[15]陈逢时，子波变换理论及其在信号处理中的应用，北京：国防工业出出版社，1998：20-162
[16]Wang DCC, Vagnucci A H, LCC. Gradient inverse weighted smoothing scheme and evaluation of its performance. 1981, CGIP 16: 167-181
[17] Tukey J.W. Exploratory data analysis. Addison-Wesley, 1977
[18] Lin Yin, Ruikang Yang,Moncef Gabbouj, Yrj? Neuvo.Weighted Median Filters: A Tutorial. IEEE Transaction on circuits and systems-Ⅱ:Analog and digital signal processing,1996,43: 157-191
[19]Jaakko Astola, Petri Haavisto, Yrj? Neuvo. Vector median filter. Proceedings of the IEEE, 1990, 78: 678-689
[20]陈贺新，非线性滤波器与数字图像处理，北京：国防工业出版社，1997，9：1-23
[21]Tao Chen, Kai-Kuang Ma,Li-Hui Chen. Tri-state median filter for image denoising. IEEE Trans. image processing, 1999,8:1834-1838
[22]D. L.Donoho and I. M. Johnstone, "Wavelet shrinkage: Asymptopia", J. R. Stat. Soc. B, 1995, 57: 301-369
[23]D. L. Donoho and I. M. Johnstone, "Ideal time-frequency denoising," Technical Report, Dept. of Statistics, Standford University ,1994
[24]D. L. Donoho and I.M. Johnstone, "Ideal spatial adaptation via wavelet shrinkage," Biometrika,1994, 81: 425-455
[25]D. L. Donoho, "De-noising by soft thresholding," IEEE Trans. Information Theory, 1995, 41: 613-627
[26] PHILIPS Corporation. SAA7111 Data Sheet [Z]. 1998
[27]Altera Corporation. ACEX 1K Programmable Logic Device Family Data Sheet. 2003. Ver3.4
[28]刘静、常丹华，ACEX 1K系列CPLD的配置，重庆工商大学学报，2003，3，20(1)：44-46
[29]何伟等，FPGA/CPLD可编程逻辑器件的在系统配置方法，2003，5，26(5)：125-128
[30]刘德良等，多分辨率图像实时采集系统的FPGA逻辑设计，电子技术应用，2003，3：69-72
[31]Kassim A A, Chua K S, Fong F K, etal. DSP-based system for real-time video communications [J]. International Journal of Imaging Systems and Technology, 1999, 10(1): 47-53
[32]TI company. TMS320C62x/C67x Technical Brief, 1998
[33]TI company. TMS320C62x/C67x CPU and Instruction Set Reference Guide, 1998
[34]IDT Corporation. IDT71V424L/S Data Sheet. 2003
[35]AMCC S5933 PCI Controller Data Book, 1996
[36]曹明，陈文正，PCI总线协议的FPGA实现及驱动设计，电子技术应用，2000，7：15-18
[37]李贵山、戚德虎，PCI局部总线开发者指南[M]，西安：西安电子科技大学出版社，1997
[38]Kamel Belkacem-Boussaid, Azeddine Beghdadi, A new image smoothing
method basdon a simple model of spatial processing in the early stages of human vision. IEEE Trans. Image Processing, 2000, 9: 220-226
[39]M. K. Mihcak, I. Kozintsev, K.Ramchandran, “Spatially adaptive statistical modeling of wavelet image coefficients and its application to denoising”, in Proc. IEEE Int. Conf. Acoust, Speech, and Signal Proc, Phoenix, AZ, March 1999, 6: 3253-3256
[40]M. K. Mih?ak, I. Kozintsev, K. Ramchandran, P. Moulin. “Low-complexity image denoising based on statistical modeling of wavelet coefficients”. IEEE Trans. on Signal Processing, December 1999, 6(12): 300-303
[41]胡昌华、张军波、夏军等，基于MATLAB的系统分析与设计－小波分析，西安：西安电子科技大学出版社，1999，12：241-246
[42]王艳辉，王永臣，小波变换技术用于图像信号的降噪，沈阳工业学院学报，2000，12，19 (4)：2-15
[43]吴小培，黄端旭，汪炳汉，图像细节保持中值滤波器，图像图形学论文集－第五届全国图像图形学学术会议，1：38-43
[44]D. L. Donoho and I. M. Johnstone, “Adapting to unknown smoothness via wavelet shrinkage”, Journal of the American Statistical Associatio,1995, 90: 1200-1224
[45]李先涛，基于小波变换的维纳滤波器图像去噪方法的研究，吉林大学硕士学位论文，2003
[46]刘刚，屈梁生，自适应阈值选择和小波消噪方法的研究，信号处理，2002，12，18(6)：509-512
[47]曲天书等，基于SURE无偏估计的自适应小波阈值去噪，电子学报，2002，2，30(2)：266-268
[48]Xiao-ping Zhang, M. Desai. “Adaptive denoising based on SURE risk”, [J]. IEEE Signal Processing Letters. October 1998, 5: 265-267
[49]Xiao-ping Zhang, M. Desai, “Nonlinear adaptive noise suppression based on wavelet transform”, In: Proceeding of the IEEE International Conference on Acoustics, and Signal Processing[C], 1997: 1589-1592
[50]G. P. Nason, “Choice of the threshold parameter in wavelet function estimation”, Wavelets and Statistics, vol.103, LNS: 261-280, New York, 1995. Springer-Verlag
[51]Weyrich N, Warhola G T, “Wavelet shrinking and generalized cross validation for image denoising”, [J]. IEEE Trans. Image Processing, 1998, 7(1): 82-90
[52]Jansen M, Malfait M, Bultheel A, “Generalized cross validation for wavelet thresholding”, Signal Processing, 1997, 56(1): 33-44
[53]谢杰成等，小波图像去噪综述，中国图象图形学报，2002，3，7(3)：209-217
[54]谢杰成等，一种小波去噪方法的几点改进，清华大学学报，2002，42(9)：1269-1272
[55]Vidakovic B, Lozoya C B, “On time-depentent wavelet denoising”, [J]. IEEE Trans. Signal Processing, 1998, 46(9): 2549-2551
[56]Ching P C, So H C, Wu S Q, “On wavelet denoising and its application to time delay estimation”, [J]. IEEE Trans. Signal Processing, 1993, 47(10): 2879-2882
[57]Shark L K, Yu C, “Denoising by optimal fuzzy thresholding in wavelet domain”, [J]. Electronics Letters, 2000, 36(6): 581-582
[58]Malfait M, Roose D, “Wavelet-based image denoising using a Markov random field a priori mode”, [J]. IEEE Trans. Image Processing, 1997, 6(4): 549-565
[59]Mallat S, “A theory for multiresolution signal decomposition: the wavelet representation”,[J]. IEEE Trans. On Pattern Analysis and Machine Intelligence, 1989, 11(7): 674-693
[60]Sweldens W, “The Lifting Scheme: a Construction of Second Generation Wavelet”, [J]. SIAM J Math Anal, 1997, 29(2): 511-546
[61]周宁等，提升小波快速算法及其在JPEG2000中的应用，中国有线电视，2002，(18)：6-10
[62]程佩青，数字信号处理教程，北京：清华大学出版社，1998：215-252
[63]虞湘宾、毕光国，长序列信号快速相关及卷积的算法研究，电路与系统学报，2001，12，6(4)：78-8
© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号 地址：北京市海淀区学院路29号 邮编：100083 电话：办公室：(+86 10)66554848；文献借阅、咨询服务、科技查新：66554700 |