摘要
波前编码技术通过在普通光学系统中加入一块特殊的相位板对波前进行调制,使得其光学传递函数和点扩散函数对离焦不敏感,来达到扩大光学系统的焦深的目的。光学系统采用波前编码技术时在探测器上得到的是模糊的中间像,需要结合有效的数字图像复原技术来得到最终的清晰像。
本文围绕适用于波前编码系统中的图像复原算法进行研究。波前编码系统的点扩散函数PSF的形式是已知的,通过软件模拟和实验都可以得到,但它与传统的运动或散焦引起的降质函数都不同,具有特殊的形式。在实际波前编码系统成像实验的基础上,对不同复原算法的效果进行了探讨。针对波前编码系统图像复原中存在的噪声放大、信噪比降低等难题,给出了具体的解决方法,并用实验进行了验证。主要研究内容包括以下几方面:
1.综合阐述了波前编码技术及其图像复原研究的历史发展和现状,同时分析了波前编码系统中图像复原的特点,以及进行此项研究的意义。
2.介绍了数字图像复原的基本理论,包括图像复原的基本概念、问题模型、经典的几种复原算法及图像质量评价方法等,并对影响图像复原效果的噪声和各种去噪方法进行了简单描述。
3.对波前编码技术的理论基础进行了研究,得到了基于模糊函数和稳相法的三次位相板形式。设计了一个三次位相板的编码系统,通过光学设计软件对其成像特性进行了分析,并进行了镜头的研制和成像实验,探讨了波前编码成像技术的成像效果和焦深扩展能力。
4.对波前编码系统成像的噪声来源和种类进行了分析,建立了系统的噪声模型和频率信噪比模型,针对波前编码系统图像复原过程中存在的噪声放大、信噪比降低问题,提出了使用一种改进的复原算法,将小波去噪引入了传统的LR复原算法中,解决了原始LR算法的噪声放大问题。实验结果表明,该方法有效地抑制了噪声,得到了很好的复原效果。
5.为了进一步提高波前编码系统的分辨率,进行了超分辨率复原的研究。提出了使用一种基于最优重构理论的复原方法,通过对波前编码系统的成像和复原模拟,发现最优重构的方法有效地改善了波前编码系统复原图像的分辨率。
By adding a special phase plate in general optical systems to modulate the wave-front, making its optical transfer function and point spread function insensitive to defocus, wave-front coding technology can extend depth of focus of optical systems. Optical systems using wave-front coding technology obtain intermediate blurred images from the detector, and an effective digital image restoration technique is combined to get the final clear image.
This dissertation focuses on the research of image restoration algorithms suited for wave-front coding systems. The point spread function (PSF) of a wave-front coding system is available and can be obtained by software simulation or experiments. It has a special form different from degradation functions caused by motion or defocus blur. Based on the imaging experiment of actual wave-front coding system, the effects of different restoration algorithms are discussed. We proposed a detailed method to solve problems such as noise amplification and signal-to-noise ratio presence in wave-front coding image restoration and the corresponding solution is verified by experiments. The main contents include the followings:
1. The dissertation describes the history and current situation of wave-front coding technology and its image restoration, and analyzes the characteristics of image restoration in the wave-front coding system, as well as the significance of the study.
2. The basic theory of digital image restoration is introduced, including the basic concept, problem model, several classic restoration algorithms and evaluation methods of the image quality in image restoration, and briefly describes the noises that impact image restoration effects and various denoising methods.
3. The dissertation researches the basic theory of wave-front coding technology, finds cubic phase plate form based on ambiguity function and stationary-phase method, and analyzes its imaging characteristics using optical design software, after fabrication of the lens, the imaging experiments are carried out, discusses imaging effects and extending the depth of focus ability of the wave-front coding technology.
4. The dissertation analyzes sources and types of noises in wave-front coding system imaging, establishes the system noise model and frequency signal to noise ratio model. In order to solve the problem of noise amplification in the restoration process, an improved algorithm is introduced. It cooperates wavelet denoising into the traditional LR algorithm. The experimental results show that the new method is effective.
5. In order to further improve the resolution of wave-front coding system, super-resolution restoration technology is studied, the dissertation exhibits an image restoration method based on the optimal reconstruction theory. Imaging experiments and restoration simulation find that the optimal reconstruction method is effective to improve the resolution of the wave-front coding system restoration image.
引文
[1]郁道银,谈恒英.工程光学[M].第2版.北京:机械工业出版社,2006.193-194.
[2] Mino M,Okano Y. Improvement in the optical transfer function of a defocused optical system through the use of shaded apertures[J]. Applied Optics,1971,10:2219–2225.
[3] Ojeda C J, Andres P and Diaz A. Annular apodiz-ers for low sensitivity to defocus and to spherical aberation[J]. Opt Lett,1986, 11:487–489.
[4] Ojeda C J and Berriel-Valdos L R. Zone plate for arbitrarily high focal depth[J].?Applied Optics, 1990,29:994–997.
[5] G Hausler. A method to increase the depth of focus by two step image processing[J]. Optics Communications,1972,6: 38–42.
[6] Brenner K H, Lohmann A, Ojeda J C. The Ambiguity Function as a Polar Display of the OTF[J].Optics Communications, 1983,44:323-326.
[7] Tucker S C,Cathey W T,and Dowski E R. Extended depth of field and aberration control for inexpensive digital microscope systems[J].Optics Express,1999, 4(11):467-474.
[8] Muyo G, Harvey A R. Wavefront coding for athermalization of infrared imaging systems [C]. Proc of SPIE, 2004, 5612: 227-235.
[9] Marks D L, Stack R A, Brady D J.Three-dimensional tomography using a cubic-phase plate extended depth-of-field system[J]. Opt Lett,1999,24(4):253-255.
[10] Narayanswamy R, Silveira P E X. Iris recognition at a distance with expanded imaging volume[C]. Proc of SPIE,2006,6202:1-12.
[11] Narayanswamy R,Johnson G E,and Silveira P E X. Extending the imaging volume for biometric iris recognition[J]. Applied Optics, 2005,44:701-712.
[12]陈燕平.景深延拓波前编码系统理论及其应用研究[D].[博士学位论文].杭州:浙江大学信息科学与工程学院,2007.
[13] Dowski E R, Kubala K. Reducing size, weight, and cost in a LWIR imaging system withWavefront Coding[C]. Proc of SPIE,2004,5407:66-73.
[14] Porras R, Vázquez S, Castro J. Wavefront coding technology in the optical design of astronomical instruments[C]. Proc of SPIE,2004,5622:796-800.
[15] Muyo G, Harvey A R. Harvey Decomposition of the optical transfer function:wavefront coding imaging systems[J]. Opt Lett,2005,30(20):2715-2717.
[16] Abrahamsson S, Usawa S and Gustafsson M. A new approach to extended focus for high-speed, high-resolution biological microscopy[C]. Proc of SPIE,2006,6090.
[17] Kubala K, Dowski E R, Cathey W T. Reducing complexity in computational imaging systems[J]. Optics Express, 2003, 11:2102-2108.
[18] Dowski E R, Jr, Johnson G E. Wavefront Coding: A modern method of achieving high performance and/or low cost imaging systems[C]. Proc of SPIE,1999,3779:137-145.
[19] Wach H B, Dowski E R, Jr, et al. Control of chromatic focal shift through wavefront coding[J]. Applied Optics, 1998, 37(23):5359-5367.
[20] Zhang W Z. Point spread function characteristics analysis of the wavefront coding system[J]. Optical Express,2007,15(4): 467-474.
[21] Dowski E R, Jr, Cormack R H, et al.Wavefront Coding: jointly optimized optical and digital imaging systems[C]. Proc of SPIE,2000,4041:114-120.
[22] Cathey W T, Dowski E R. New paradigm for imaging systems[J]. Applied Optics, 2002, 41:6080-6092.
[23] Morales F G, Tudela R, Borrull J F,et al. Pupil filters for wavefront coding: off axis performance[C]. Proc of SPIE,2005,5962:371-377.
[24] Gracht J, Dowski E R, Jr, et al. Broadband behavior of an optical–digital focus-invariant system[J]. Opt Lett,1996,21(13):919-921.
[25] Muyo G, Harvey A R. Wavefront coding for athermalization of infrared imaging systems [C]. Proc of SPIE, 2004, 5612: 227-235.
[26] Dowski E R, Jr, Kubala K S. Modeling of wavefront coded imaging systems[C]. Proc of SPIE,2002,4736:116-126.
[27] Dowski E R, Jr,Cathey T W. Extended depth of field through wave-front coding[J]. Applied Optics, 1995, 34(11):1859-1866.
[28] Bradburn S, Cathey W T, Dowski E R, et al. Realization of invariance in optical-digital systems with wave-front coding[J]. Applied Optics,1997,36(35):9157-9166.
[29] Mezouari S, Harvey A R. Wavefront coding for aberration compensation in thermal imaging systems[C]. Proc of SPIE,2001,4442:34-42.
[30] Plemmons R,Horvath M and Leonhardt E. Computational imaging systems for iris recognition[C], Proc of SPIE,2004,5559,346-357.
[31] Morales F G, Tudela R, Borrull J F,et al. Pupil filters for wavefront coding: off axis performance[C]. Proc of SPIE,2005,5962:371-377.
[32] Yan Feng,? Zheng Li-gong, Zhang Xue-jun. Image restoration of an off-axis three-mirror anastigmatic optical system with wavefront coding technology[J].? Optical Engineering,2008,47(1):1-8.
[33] Abrahamsson S, Usawa S and Gustafsson M. A new approach to extended focus for high-speed, high-resolution biological microscopy[C]. Proc of SPIE,2006,6090.
[34] Gracht J,Nagy J,Pauca P, et al. Iterative restoration of wavefront coded imagery for focus invariance,”in OSA Trends in Optics and Photonics (TOPS), Integrated Computational Imaging Systems, OSA Technical Digest, Washington, D.C., 2001.
[35] Johnson G E, Macon A K, Rauker G M.? Computational imaging design tools and methods[C]. Proc of SPIE,2004,5524,284-294.
[36]邓泽峰.图像复原技术研究及应用[D].[博士学位论文].武汉:华中科技大学机械电子工程,2007.
[37] Gonzalez R C, Woods R E. Digital image processing, second version[M].北京:电子工业出版社.2002.220-230.
[38]李玉峰.小波分析在图像去噪与压缩中的应用研究[D].[博士学位论文].长春:中国科学院长春光学精密机械与物理研究所,2006.
[39] Mallat S. A theory for multiresolution signal decomposition: the wavelet representation[J]. IEEE Trans Pattern Analysis and Machine Intelligence 1989,11:674-693.
[40] Mallat S, Zhong S. Characterization of signal from multiscale edges[J]. IEEE Trans Pattern Analysis and Machine Intelligence, 1992,14:710-732.
[41] Mallat S, Hwang W L. Singularity detection and processing with wavelets [J]. IEEE TransInformation Theory ,1992,38(2):617-643.
[42] Xu Yansun, Weaver J B, Healy M J,et al. Wavelet transform domain filters: a spatially selective noise filtration technique [J]. IEEE Trans Image Processing, l994, 3(6):747-758.
[43] Donoho D L, De-noising by soft-thrsholding[C], IEEE Trans. Information Theory,1995,141:613-627.
[44] Donoho D L,Johnstone I M, Kerkyacharian G. Wavelet shrinkage: Asymptopia[J] Roy Stat Soc ser B, 1995,57:301-369.
[45] Donoho D L. Nonlinear wavelet methods for recovery signals, densities, and spectra from indirect and noisy data[J],Proc of Symp in Appl Math ,1993,100, 173-205.
[46] Donoho D L , and Johnstone I M .Ideal spatial adaptation via wavelet shrinkage, Biometrika[J], 1994,18:425-455.
[47] Donoho D L, Johnstone L M. Adapting to unknown smoothness via wavelet shrinkage[J].Journal of the American Statistical Association,1995, 90(432):1200-1224.
[48] Gao H. Wavelet shrinkage denoising using the non-negative garrote[J]. Journal of Computational and Graphical Statistics, 1998,7(4): P469-488
[49] Gao H. and Bruce A. WaveShrink with firm shrinkage[J]. Statistica Sinica, 1997, 7: 855-874.
[50] Bruce A. and Gao H. waveShrink: Shrinkage function and thresholds[J]. SPIE, 1996,2569: 270-281.
[51] Johnstone I M, Silverman B W. Wavelet threshold estimators for data with correlated noise[R].Department of statistics, Stanford University, 1992.
[52] Jansen M, Bultheel A. Multiple wavelet threshold estimation by generalized cross?validation for Images with correlated noise[J]. IEEE Trans. on IP, 1999,78(7): 947-953.
[53] Satio N. Simultaneous noise suppression and signal compression using a library of orthonormal bases and the minimum description length criterion[J].in Wavelet in Geophysics,?1994, 299-324.
[54] Hsung T C. Denoising by singularity detection[J]. IEEE Trans. Signal Processing, 1999,47(11):3139-3144.
[55] Chang S G, Yu B, Verserli M. Spatially adaptive wavelet thresholding with context modeling for image denoising[J].IEEE Trans Image Processing,2000,9(9):1522-1530.
[56] Chang S G, Yu B, Vetterli M. Adaptive wavelet thresholding for image denoising and compression[J].IEEE Trans Image Processing, 2000,109:1532-1546.
[57] Starck J L. The curvelet transform for image denoising [J]. IEEE Trans Image Processing, 2002,11(6):670-684.
[58] Nason G P, Silveman B W. The stationary wavelet transform and some statistical applications in wavelet and statistics[M]. A Antoniadis ed Lecture Notes in statistics Berlin, Spinger Verlag 1995:281-299.
[59]肖小奎,黎绍发.加强边缘保护的Curvelet图像去噪方法[J].通信学报,2004.
[60] Strela V, He11er P N. The application of multiwavelet filterbanks to image processing [J].IEEE Trans Image Processing,1999,8(4):548-563.
[61] Xia X G. A new prefilters design for discrete multiwavelet transforms[J]. IEEE Trans Signal Processing,1998,46(6): 1558-1570.
[62]费佩燕,郭宝龙.基于多小波的图像去噪技术研究[J].中国图象图形学报,2005, l0(1):107-112.
[63]周岐,值小光,周军.基于平稳多小波变换的红外图像噪声抑制方法[J].红外与毫米波学报,2005,24(2):151-155.
[64]长光新,崔扬,周泽魁.基于小波包分解的纹理图像去噪[J].华南理工大学学报(自然科学版),2005,33(3):31-49.
[65]张毅刚,郁惟墉等,基于小波包及隐式马尔科夫模型的局放信号去噪[J],上海交通大学学报,2004,38(8):1269-1272.
[66]范显峰,姜兴渭.基于小波包变换的信号去噪方法研究[J],哈尔滨工业大学学报,2003, 35(7):809-811.
[67]谢义方.数字图像复原算法研究[D].[硕士学位论文].长沙:湖南大学电气与信息工程学院,2004.
[68]吴亚东.图像复原算法研究[D].[博士学位论文].成都:电子科技大学计算机应用技术,2006.
[69] Ozkan M K, Erdem A T, Sezan M L, et al. Efficient multiframe Wiener restoration ofblurred and noisy image sequences[J]. IEEE Transactions on Image Processing, 1992,1(4): 453-476.
[70] Wu W, Kundu A. Image estimation using fast modified reduced update Kalman filter[J]. IEEE Trtmsactions on Signal processing, 1992,40(4): 915-926.
[71] Citrin S, Azimi-Sadjadi M R.A full-plane block Kalman filter for image restoration[C]. IEEE Transactions on Image Processing,1992,1(4):488-495.
[72] Koch S, Kaufinan H, Biemond J. Restoration of spatially varying blurred images using multiple model-based extended Kalman filters[J]. IEEE Transactions on Image Processing, 1995.
[73] Andrews H C, Hunt B R.Digital image restoration[M]. Englewood Cliffs: Prentice-Hall,1977,113-116.
[74] Hurnt B R. The application of constrained least squares estimation to image restoration by digital computer[J]. IEEE Transactions on Computers, 1973, C-22 (9): 805-812.
[75] Erler K, Jernigan E. Adaptive image restoration using recursive image filters[J]. IEEE Transactions on Signal Processing,1994,42(7):1877-1881.
[76] Mallikarjuna H S, Chaparro L F.Iterative composite filtering for image restoration[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1992,14(6):674-678.
[77] Kuzdur D, Hatzinakos D. A novel blind deconvolution scheme for image restoration using recursive filtering[J]. IEEE Transactions on Signal Processing, 1998,46(2):375-390.
[78] Chow T W S, Li X D, Cho S Y. Improved blind image restoration scheme using recurrent filtering[J]. IEEE Proceedings-Vision, Image and Signal Processing, 2000,147(1):23-28.
[79] Gokmen M, Li C. Edge detection and surface reconstruction using refined regularization[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1993,15(5):492-499.
[80] Zervakis M E, Venetsanopoulos A N. Iterative least squares estimators in nonlinear image restoration[J]. IEEE Transactions Signal Processing,1992,40(4): 927-945.
[81] Rafael C G,Richard E W,Digital Image Processing[M].Beijing: Publishing House of Electronics Industry, 2002, 231-234.
[82] Wendt P D, Coyle E J, Gallagher N C J. Stack filters[J]. IEEE Trans on Acoust Speech Signal Processing,1986,34(4):898-911.
[83] Serra J. Image analysis and mathematical morphology[M]. New York: Academic Press,1982:34-40.
[84]蔡靖,杨晋生,丁润涛.模糊加权均值滤波器[J].中国图像图形学报,2000, 5(A) (1):52-56.
[85]王伟,赵春辉,孙圣和.缓优全方位结构元约束二维层叠滤波器[J].中国图像图形学报,1999, 4(A)(6): 445-449.
[86]赵春辉,王伟.全方位多结构元形态滤波器[J].中国图像图形学报,1997, 2(7): 500-503.
[87] Eng H L, Ma K K. Noise adaptive soft-switching median-filter[J].IEEE Transaction on Image Processing,2001,10(2):242-251.
[88] Gouchol P, Liu J C, Nair A S. Selective removal of impulse noise based on homogeneity level information[J]. IEEE Transaction on image Processing, 2003, 12(1): 85-92.
[89] Castro A, Ojeda C J. Increased depth of field with phase-only filters: ambiguity function[C]. Proc of SPIE,2005,5827:1-11.
[90] Bagheria S,Silveirab P E X,and Narayanswamy R. Design and optimization of the cubic-phase pupil for the extension of the depth of field of task-based imaging systems[C]. Proc of SPIE,2006,6311.
[91] FitzGerrell A R, Dowski E D, Jr,et al.Defocus transfer function for circularly symmetric pupils[J]. Applied Optics,1997,36(23):5796-5804.
[92] Josephson G E, Silveira P E X, Dowski E R. Analysis tools for computational imaging systems[C]. Proc of SPIE, 2005,5817:34-44.
[93] Mezouari S,Harvey A R. Phase pupil functions for reduction of defocus and spherical aberrations[J]. Opt Lett,2003,28(10):771-773.
[94] Dowski E R, Kubala K.Design and optimization of computational imaging systems[C]. Proc of SPIE,2004,5299:155-162.
[95]雷广智,张新,张建萍等.波前编码系统的新设计[J].光学精密工程,2008,16(7):1171-1176.
[96]雷广智.扩大光学系统焦深的波前编码技术研究[D].[硕士学位论文].长春:中国科学院长春光学精密机械与物理研究所,2008.
[97] Wach H B, Dowski E R, Jr. Noise modeling for design and simulation of computational imaging systems[C]. Proc of SPIE,2004,5438:159-170.
[98] Silveira P E X,Narayanswamy R. Signal-to-noise analysis of task-based imaging systems with defocus[J]. Applied Optics,2006,45(13):2924-2934.
[99]杨皓明,张新,方志良等.三次位相板编码系统的频率信噪比分析[J].光子学报,2008,37(7):1412-1415.
[100] Liu J,Moulin P. Image denoising based on scale-space mixture modeling of wavelet coefficients[J].Proceedings of IEEE international Conference on Image Processing, 1999:386-390.
[101] Lang M, Guo H, Odegard J ,et al.Noise reduction using an undecimated discrete wavelet transform[J]. IEEE Signal Processing Letters 1995,3(1):10-12.
[102] Crouse M S, Nowak R D, and Baraniuk R G. Wavelet-based signal processing using hidden Markov models[J]. IEEE Trans Signal Processing,1998,46(4):886-902.
[103] Wu C B, Liu B D, Yang J F. A fuzzy-based impulse noise detection and cancellation for real-time processing in video receivers[J].IEEE Trans Instrumentation and measurement,2003, 52(3):780-784.
[104] Zhong J M, Ning R L,Conover D.Image denoising based on multiscale singularity detection for cone beam CT breast imaging[J].IEEE Trans Medical Imaging,2004, 23(6):696-703.
[105] Berkner K, Wells R O. A correlation-dependent model for denoising via nonorthogonal wavelet transforms[R], Rice University, Houston,1998.
[106] Satio N. Simultaneous noise suppression and signal compression using a library of orthonormal bases and the minimum description length criterion in Wavelet in Geophysics[M], New York, 1994, 299-324.
[107]金炜,魏彪.中子数字图像几何不锐度校正算法研究[J].光学学报,2007,27(10):1765-1770.
[108]李盛阳.图像盲复原与改善空间分辨率研究[D].[博士学位论文].北京:中国科学院遥感应用研究所,2006.