This action might not be possible to undo. Are you sure you want to continue?

University of Alicante, Valencia, Spain, Dec. 21, 2004

Digital holography and image processing: twins born by the computer era Digital holography: - computer synthesis, analysis and simulation of wave fields Digital image processing: - digital image formation; - image perfection; - image enhancement for visual analysis; - image measurements and parameter estimation; - image storage & transmission; - image visualization

•Digital reconstruction of of electronically recorded optical holograms •Target location and tracking •Multi-component image restoration •Image re-sampling with discrete sincinterpolation •Nonlinear (rank) filters for image restoration, enhancement and segmentation •3-D visualization

DIGITAL HOLOGRAPHY: numerical reconstruction of electronically recorded optical holograms

Digital Holographic/Interferometric Microscopy

One of the main drawbacks of microscopy: the higher is the spatial resolution, the lower is depth of focus. This problem can be resolved by holography. Holography is capable of recording 3-D information. Optical reconstruction is then possible with visual 3-D observation. Drawbacks of optical holography: -Intermediate step (photographic development of holograms) is needed. -Quantitative 3-D analysis requires bringing in additional facilities Radical solution: optical holography with hologram recording by electron means (digital photographic cameras) and digital reconstruction of holograms. This is the principle of digital holographic microscopy.

Beam spatial filter Collimator Lens Digital Photographic camera

Laser Object table

Microscope

Computer

Digital Reconstruction of Holograms (Equivalent optical setup) Hologram First focal plane Fourier Plane Second focal plane .

v.S.P. Reconstruction of a Hologram with a Computer. p. Yaroslavsky. 2. N. 419 . 1972. Merzlyakov. 17. Soviet Physics-Technical Physics.Digital Holography: Digital Reconstruction of Holograms M. Kronrod.420 Hologram Hologram sensor Computer Analog-todigital conversion Preprocessing of digitized hologram Image reconstruction (DFT/DFrT) Image processing Output image . no. L.A.

PSF. resolving power and speckle phenomena in reconstruction of electronically recorded optical holograms Problems: •How the point spread function of the reconstruction process depends on parameters of the optical set-up and recording camera •Resolving power of the reconstruction process •Potential accuracy in measuring phase component of the object wavefront •Statistical properties of the reconstruction speckle noise .

Yaroslavsky. and A. 1980.P. Cappellini. Shifted Discrete Fourier Transforms. 69.v = 1 N N −1 k =0 ∑a k ru ⎞ k (r + v ) ⎞ ⎛ ⎛ exp⎜ − i 2π ⎟ exp⎜ i 2π ⎟ N⎠ N ⎠ ⎝ ⎝ .v r 1 = N ∑ ak exp⎜ i 2π (k + u )(r + v ) ⎞ ⎛ ⎟ N ⎝ ⎠ N = 1 / Δ xΔ f Direct and inverse Shifted DFTs (reduced form) α ru . by V. Sampled signal Δx x a( x ) = N −1 k =0 ∑ akϕ sign _ reconstr ( x − (k + u )Δx ) Sampled signal spectrum Continuous signal spectrum α( f ) v Δf α( f ) = f N −1 k =0 ∑ α rϕ spn _ reconstr ( f − (r + v )Δf ) Shifted DFT (canonic form) N −1 k =0 Fourier integral ∞ Signal and spectrum sampling α ( f ) = ∫ a ( x )exp(i 2πfx )dx −∞ α u . Ed. London. G. In: Digital Signal Processing.74. p.v = 1 (k + u )r ⎞ kv ⎞ ⎛ ⎛ ∑ a k exp⎜ i 2π N ⎟ exp⎜ i 2π N ⎟ N k =0 ⎝ ⎠ ⎝ ⎠ N −1 α ru . Constantinides. Avademic Press.Discrete representation of optical transforms: Discrete Fourier Transforms Continuos signal a(x) u L.

r ) = 1 0.3 0. Horizontal coordinateN– qargument N=256 and function. q. .0001 {χ r ( f ) = sinc[π ( f − (r + v )Δf ) / Δf ]} Discrete Fresnel Transforms α rκ .0004 q=0. w = κ ak .6 0. Amplitude is shown in green.0016 ∑ α κ exp{iπ (kκ − r / κ + w ) r /N } N = 1 / ΔxΔfD 2 κ = (Δf / Δx ) 1/ 2 w is the overall shift w=uκ-v/κ 50 100 200 A 0pictorial representation of the 150 discrete frinc-function 250 (DFT of the chirp-function). r for of the Absolutein red.1 0 q=0.9 2005 1 N N −1 k =0 kr ⎞ ⎛ ∑ exp(iπqk )exp⎜ − i 2π N ⎟ ⎠ ⎝ 2 {φ k ( x ) = sinc[π ( x − (k + u)Δx ) / Δx ]} Signal and its transform discretization with shift basis functions q=0 0.5 q=0.0008 q=0.w = 1 N 1 N N −1 k =0 ∑a N −1 k =0 k exp − iπ (kκ − r / κ + w ) / N 2 2 { } 0.0002 q=0. Binary different “focusing” parameter q random noise is added just for fun.7 0. vertical coordinate – focusing parameter.4 0.2 0.Discrete representation of optical transforms: Discrete Fresnel Transforms Integral Fresnel Transform: ⎡ ( x − f )2 ⎤ dx α ( f ) = ∫ a ( x )exp⎢ − iπ ⎥ D ⎦ −∞ ⎣ ∞ frinc ( N . ( ) . phase values of function frinc .8 0.

Point spread function of numerical reconstruction of electronically recorded optical holograms Digital reconstruction of samples of the object wave front amplitude from samples of its hologram is treated as a process of sampling the object wave front. ideal sampling PSF is sinc-function PSF (k . x ) = sinc[π ( x − kΔx ) Δx ] sin[π ( x − kΔx ) Δx ] = [π ( x − kΔx ) Δx ] . Signal sampling is a linear transformation that is fully specified by its point spread function: ak = ∫ a ( x )PSF ( x . k )dx X Wave front samples Object’s wave front Sampling device Point Spread Function According to the sampling theorem.

8-12 November 2004. L. k ) = Φ d ( Define hologram discretization and reconstruction device coordinate system through the object coordinate system by choosing vr=vT=(N-1)/2. Yamaguchi.f) is : WP ( x . for hologram reconstruction. shifted and scaled DFT is used with the reconstruction kernel: DR( k . k ) = Φ d ( ⎡ ⎤ x Δx ⎞ ⎛ ) sincd ⎢ N r . π ⎜ x − k ⎟ Δx ⎥ σ ⎠ λZ ⎝ ⎣ ⎦ .PSF of reconstruction of holograms recorded in far diffraction zone spread functions of digital reconstruction of digitally recorded holograms. Point In: Photonics Asia conference on Information Optics and Photonics Technology. Then PSF FZ ( x . wave propagation kernel WP(x. π ⎜ x − k ⎟ − ⎜ vT + ⎟ Δx ⎥ σ ⎠ λZ 2 ⎠ σ⎝ 2 ⎠⎦ ⎭ ⎝ ⎣ ⎦ ⎣ Δx ⎝ ⎩ ∞ xf ⎞ x ⎛ ) ⎟df is frequency response of the hologram sampling device and where Φ d ( ) = ∫ ϕ d ( f ) exp⎜ ( − i 2π λZ λZ ⎠ ⎝ −∞ sin( x ) Δx = λ Z S H = λZ NΔf sincd( N . Zhang.k) is ⎧ ⎡ x ⎛ ⎡ ⎤ x N −1⎞ k ⎛ N − 1 ⎞⎤ ⎫ Δx ⎞ ⎛ ) exp⎨ − i 2π ⎢ ⎜ v r + ⎟ ⎥ ⎬ sincd ⎢ N . x ) = N sin( x N ) PSF FZ ( x . F. point spread function of the reconstruction process PSFFZ(x. I. r ) = 1 k (r + vT ) ⎤ ⎡ exp⎢ i 2π N σN ⎥ ⎣ ⎦ where vT and σ are shift and scale parameters With this reconstruction kernel. Beijing. Yaroslavsky. China For hologram recording in far diffraction zone. f ) = exp(− i 2π xf ) λZ Assume that.

The case σ =1 corresponds to a “cardinal” reconstructed object wavefront sampled with discretization interval Δx= λZ/ SH =λZ/ NΔf . π ( x − kΔx ) Δx ] λZ The point spread function is a periodical function of k: PSF FZ (k + gσN r ) = (− 1) g ( N r −1 ) PSF FZ (k ). One can show that in this case the reconstructed object wavefront is a discrete sinc-interpolated version of the “cardinal” one.PSF of reconstruction of holograms recorded in far diffraction zone (ctnd) As one can see from the equation. (g is integer). . It generates σN samples of object wavefront masked by the frequency response of the hologram recording and sampling device. When σ >1 . the samples being taken with discretization interval Δx/σ = λZ/ σSH =λZ/ σNΔf within the object size So= λZ/ Δf. reconstructed discrete wavefront is σ -times over-sampled. PSF FZ ( x . or σ -times zoomed-in. k ) = Φ d ( x ) sincd[N r .

As distinct from the sinc-function.Discrete sinc-function sin(πx Δx ) sincd( N . which is a point spread function of the ideal low-pass filter. discrete sinc-function is a periodical function with period NΔx or 2NΔx depending on whether N is an odd or an even number and its Fourier spectrum is a sampled version of the frequency response of the ideal low pass filter N is an odd number N is an even number NΔx 2NΔx Reconstructed object area Continuous (red dots) and discrete (blue line) sinc-functions for odd and even number of samples N Frequency response of the ideal low pass filter (red) and Fourier transform of the discrete sinc-function (blue) Reconstructed object area . πx Δx ) = N sin( x NΔx ) Discrete sinc-function is a discrete analog of the continuous sampling sinc-function.

original image. Statistical characterization of speckle noise in coherent imaging systems. in: Optical Measurement Systems for Industrial Inspection III. Germany. pp.image reconstructed in far diffraction zone from 0. A. .image reconstructed in far diffraction zone from 0. Munich. K. Osten. b) . c) . 23-25 June 2003. Shefler. Eds. W.5 of area of the wave front. Yaroslavs ky. SPIE’s Int.Statistical characterization of speckle noise in coherent imaging L. 175-182 a) b) Generating 2-D array of pseudo-random numbers that specify the phase component of the object wave front Computing object’s wave front Simulating wave front propagation (DFT.9 of area of the wave front. Kujawinska. SPIE v. IDFrT) Computer model Illustrative examples of simulated images: a) . DFrT) Introducing signal distortions: -Array size limitation -Dynamic range limitation -Quantization 2-D array that specifies amplitude component of the object wave front c) d) Yes Continue iterations ? No Comparing reconstructed and initial wave fronts. computing and accumulation of noise statistical parameters Output data Simulating wave front reconstruction (IDFT. Creath. Symposium on Optical Metrology. d) image reconstructed in far diffraction zone after limitation of the wave front orthogonal components in the range.. 5144. M.

TARGET LOCATION AND TRACKING .

No 3 (1999) 355-369 . 8. Vol. Yaroslavsky. Optimal target location in color and multi component images.Target localization in clutter in multi-component images L. Asian Journal of Physics.

ac. Ben-Zion Shaick Transform Oriented Image Processing Technology for Quantitative Analysis of Fetal Movements in Ultrasound Image Sequences.il/~yaro Tracking fetus movements in Ultrasound movie . Theodorisdis. I. 1998. Kalouptsidis. 1998.Object tracking in video sequencies: examples Leonid P. 1745-1748 For details see http://www.eng.. Greece. A. Yaroslavsky.tau. By S. Typorama Editions. Theories and Applications. p. N. In: Signal Processing IX. Stouraitis. Rhodes. ed. Proceedings of Eusipco-98. Pitas. 8-11 Sept.

Vision.il/~yaro/Zion/ZionPhdSeminar4Web. The face detection capability of the developed system was experimentally examined on two test databases of images of high and low quality. Object Localization Using Linear Adaptive Filters. November 21-23. Stuttgart.Face detection in complex images Ben-Zion Shaick. GermanyStuttgart. . L. Germany. 6th Fall Workshop. November 21-23. 11-17 http://www. with high reliability. faces of varying size from minimum size of 12 pixels width and 15 pixels height to the maximum size of the input image. respectively.ac. 2001. The detection rates 96% and 84% were achieved for these databases. Yaroslavsky. pp.eng.pdf The developed algorithm is capable of detecting. 2001.tau. Modeling And Visualization 2001 (Vmv01).

. thus leaving only 1% of the image area for subsequent thorough analysis by the “multitemplate classification” algorithm. The algorithm is fast and requires approximately 200 flops per pixel in an input image of 640480 pixels size.3% (faces wrongly rejected).Face detection: two-stage algorithm 1st stage Face Non-face Maybe face 2-nd stage Facelikenonface Nonface likeface The “non-face” detection algorithm was proved to have “non-face” rejecting rate of ~99% and false alarm rate of 1.

000 images and one “non-face” database to one million images by means of scaling and rotating database images. The developed algorithms were trained using a specially created training database obtained by extending four “face” databases to 32. In particular. “face”. “faces like clutter” and “clutter like faces” templates were generated from these training databases. “nonface”.Face detection: Face-like-non-face and non-face-like face data bases “Multi-template classification” algorithms use a very large set of templates prepared for different target shapes and varying illumination conditions. .

MULTI COMPONENT IMAGE RESTORATION .

enhancement and target location with local adaptive filters. ICOIV.Asakura. 1999. 111-127 . by T. pp. Image Restoration.Spatial/temporal adaptive linear filters L. ed. in:International Trends in Optics and Photonics. Yaroslavsky. Springer Verlag.

3-D Local adaptive spatial-temporal filtering: denoising and deblurring thermal video (ctnd) .

Stockholm. A.3-D Local adaptive spatial-temporal filtering: denoising and deblurring thermal video L. Sh.1 July 2004 . Processing and Fusion of Thermal and Video Sequences for Terrestrial Long Distance Observation Systems Processing and Fusion of Thermal and Video Sequences for Terrestrial Long Distance Observation Systems. Sweden. Before After (5x5x5 DCT domain filtering) . 28 June . Seventh International Conference on Information Fusion (FUSION 2004). Gepstein. Stainman. ISIF. Yaroslavsky. B. Fishbain.

IMAGE RESAMPLING: discrete sinc-interpolation .

Image resampling and geometrical transformations: efficient discrete sinc-interpolation algorithms .

Spline interpolation and discrete sinc-interpolation Spectrum of DFT-error (green) and bicubic-error (red) Test random image S pe c tra o f DFT-e rro r (le ft) and Beta11-e rro r (rig ht) Spectrum of DFT-error (red) and Beta11 error (green) .

d) – nearest neighbor interpolation. c) – bicubic interpolation. b) – bilinear interpolation.Discrete sinc-interpolation for image resizing Image (a) is reduced with reduction factor 77/256 and then magnified with magnification factor 256/77. e) – sincd-interpolation 11x11. f) – sincd interpolation 31x31 .

. p.Discrete sinc-interpolation in DCT domain L. 10 July 2003. Yaroslavsky.4166-4175 Initial signal (ak ) DCT αrDCT DFT of zero-pad sincinterpolation kernel re r {η {α } DCT r ( p )} {ηr ( p )} {η im r ( p )} IDCT + IDST - p-shifted sinc-interpolated signal {ak + p } Zooming an image fragment (left) by sinc-interpolation in DFT domain (right upper image) and in DCT domain (right bottom image). v. 42. 20. Oscillations due to boundary effects that are clearly seen in DFT-interpolated image completely disappear in DCT-interpolated image. Boundary effect free and adaptive discrete sinc-interpolatioon. No. Applied Optics.

Sliding window sinc-interpolation DCT domain: Simultaneous image resampling and restoration/enhancement Input signal Inverse DCT/DST Output signal Computing sliding window DCT Introducing p-shift Modification of the spectrum for restoration/ enhancement ∑ .

Sliding window sinc-interpolation in DCT domain: signal resampling and denoising a) b) Noisy image (a) and a result of the rotation and denoising with sliding window DCT sinc-interpolation and denoising (b). .

Sliding window sinc-interpolation DCT domain: local adaptive interpolation Discrete DCT domain sinc-interpolator Input signal Computing sliding window DCT Analysis of local spectrum Mixer Output signal Nearest neighbor interpolator .

from top to bottom: Continuous signal. linearly interpolated signal. bicubic spline and adaptive sliding window sinc interpolation methods for zooming a digital signal (From left to right. cubic spline -interpolated signal. . initial sampled signal. sliding window sinc-interpolated signal ). Comparison of nearest neighbor. One can notice disappearance of oscillations at the edges of rectangle impulses when interpolation is adaptive.Adaptive versus non adaptive signal interpolation Signal (upper plot) shift by non-adaptive (middle plot) and adaptive (bottom plot) sliding window DCT sinc-interpolation. linear. nearest neighbor -interpolated signal .

Image rotation with adaptive and non-adaptive discrete sinc interpolation .

Cartesian coordinate spectrum conversion with discrete sinc-interpolation Object Projections Projection spectra Interpolated 2-D object spectrum Reconstructed image .Direct Fourier method for inverse Radon transform: polar-to.

Chernobrodov. P. Y. Italy) radius Initial Polar Grid angle 1-D spectra of projections Target Cartesian Grid Polar Grid after 2 time zooming Initial Polar Grid Target Cartesian Grid a r1 r2 b ?? . Rome.Cartesian coordinate spectrum conversion by means of zooming with variable zooming factor (L. 3-d Int. Symposium. Sept. 18-20. 2003. Yaroslavsky. Image and Signal Processing and Analysis.Fourier method for inverse Radon transform: polar-to. Sinc-interpolation methods for Direct Fourier Tomographic Reconstruction.

2004 Restored stabilized video with moving objects unaffected Moving objects Initial video . Sept. Restoration of atmospheric turbulent video containing real motion using rank filtering and elastic image registration.P. Eusipco2004. Yaroslavsky. Gepshtein.Stabilization and restoration of atmospheric turbulent video S. and L. Shtainman. Vienna. B. A. Fishbain. Austria.

Turbulent atmosphere video .

Stabilized turbulent atmosphere video .

NONLINEAR (RANK) FILTERS for image restoration. enhancement and segmentation .

Yaroslavsky. EUSPCO 2004. Pn=0. No. stdev = 20. Vienna.Image restoration and enhancement: nonlinear filters L. 12. 5x5. 2003 B. L. Nonlinear Filters for Image Processing in Neuromorphic Parallel Networks. Yaroslavsky. Optical Memory and Neural Networks. Evpl=Evmn=15. Wind. Hirshl. Austria.15 Iterative SCSigma-filter . 1. FPGA implementation of sorters for non-linear filters. 2004 Noisy image. 5 iterations . v. Sept.

2.Nonlinear filters: Image enhancement Initial image SIZE(Evnbh(Wnbh5x5.2))-filter HIST(W-nbh)-filter .

EV-nbh and KNV-nbh Initial image RANK(Wnbh15x15) RANK(KNV (Wnbh15x15.113)) RANK(EV (Wnbh15x15.Nonlinear Filters: Image Enhancement Local histogram equalization: Wnbh.10)) .10.

Local P-histogram equalization: color images (blind calibration of CCD-camera images) .

3-D VISUALIZATION: improved anaglyph method and 3-D video from 2-D video .

J. ISSN: 1482-7921) http://www.tau. Anaheim.Computer synthesis and display of stereoscopic images I. Visualization.html Bahai garden. Imaging and Image Processing (VIIP 2002). Acta Press. Ed. 2002. 191-194 (ISBN: 0-88986-354-3. Haifa. Conference. 2nd IASTED Internat.eng. Sept.J. Spain. Yaroslavsky. Ideses. Malaga. Zurich. 9-12. Calgary. 2002. Israel . L. Villanueva.il/~yaro/Ideses/malca1.ac. pp. Efficient Compression and Synthesis of Stereoscopic Video.

Roman Querry. Tarragona. Catalonia .

B.). H. Academishe Verlagsgesellschaft Aka GmbH. Greiner. ISBN 158603-302-6. Berlin. Seidel (Eds. L. Yaroslavsky. T. Proceedings.Computer generated stereo from 2-D video I. H. Vision Modeling and Visualization 2002. A Method for Generating 3D Video from a Single Video Stream. Erlangen. 2002. Girod. Ideses. 2002. 20-22. Ertl. ISBN 3-89838-034-3. 435-439 .-P. Germany. in: G. Niemann. Nov. p.

Current projects: Multi-component image processing: •Restoration and fusion of atmospheric turbulent. thermal and visible range video •Super-resolution from video sequencies •3-D color display and artificial 3-D from video •Moving object detection and tracking Digital holography: •Point spread functions and resolving power of digital reconstruction of near and far zone holograms •Theory and new algorithms for fast transforms .

.

Tel Aviv. of Interdisciplinary Studies. Ph.tau. Professor Dept. Yaroslavsky. Tel Aviv University.eng. Sc.L.il/~yaro . Faculty of Engineering.ac.D.. Israel www. Dr. Phys&Math.

- Letter-10 ways to make your PhD experience easier and more enjoyable.pdf
- Lista de Exercícios 3_Tópicos de Física Aplicada
- Lista de Exercícios 2
- Lista de Exercícios 1
- Lista 3
- Lista 2
- Lista 1
- 8. Dinâmica de Partículas Relativísticas
- 7. Radiação de Sistemas Simples
- 6. Radiação de Cargas Em Movimento
- 5. Teoria Da Relatividade Restrita
- 4. Guias de Onda e Cavidades Ressonantes
- 3. Ondas Eletromagnéticas Planas
- 2. Campos e Leis de Conservacao
- 1. Introduction
- Jackson (2ed. PT)
- Jackson Solutions (Recomendable)
- Physics - Quantum Mechanics - J. J. Sakurai - Modern Quantum
- Sakurai Solution Espanhol
- Notas de Mecanica Quantica_Capitulo 5_Sakurai
- Notas de Mecanica Quantica_Capitulo 4_Sakurai
- Notas de Mecanica Quantica_Capitulo 3_Sakurai
- Notas de Mecanica Quantica_Capitulo 2_Sakurai
- Notas de Mecanica Quantica_Capitulo 1_Sakurai
- Potencial periódico_Kroning Penney

Sign up to vote on this title

UsefulNot useful- New Earth System Online and Fully Functional!
- et4
- 460 Presentation
- 01 Basis Expansions
- IJER_2014_219
- Kalaw v Relova
- 370192 A
- Introduction to FFT
- Project
- 3Martinez
- PH2111 Engg. Physics Jan 2010 Ques. Pape r
- Universal Send Kit b1 Ip
- Lec-13
- Cordic Based Fast Radix-2 Dct Algorithm
- Ajero vs. CA
- Intro to Video1
- Instrumentation Tutorial
- Field Wiring and Noise Considerations for Analog Signals
- NI Tutorial
- Traductor Sineax m561 m562 m563
- b90.2940.0en
- PML630 Compact Load Shedding Solution
- Discreat Fouries Transform
- FY22xxS系列用户手册V2_0
- Blind Identification of Multichannel Systems Driven by Impulsive Signals
- Unit4-MSH
- OFDM for Broadband Communication
- Fundamental of Time Gated Measurement
- 58931
- Sol 04
- Digital Holography Ad Digital Image Processing