Probabilistic Model of Error in Fixed-Point Arithmetic Gaussian Pyramid

Size: px
Start display at page:

Download "Probabilistic Model of Error in Fixed-Point Arithmetic Gaussian Pyramid"

Transcription

1 Probabilistic Model o Error in Fixed-Point Arithmetic Gaussian Pyramid Antoine Méler John A. Ruiz-Hernandez James L. Crowley INRIA Grenoble - Rhône-Alpes 655 avenue de l Europe Saint Ismier Cedex France {Antoine.Meler,John-Alexander.Ruiz-Hernandez,James.Crowley}@inrialpes.r Abstract The hal-octave Gaussian pyramid is an important tool in computer vision and image processing. The existence o a ast algorithm with linear computational complexity makes it easible to implement the hal-octave Gaussian pyramid in embedded computing systems using only integer arithmetic. However, the use o repeated convolutions using integer coeicients imposes limits on the minimum number o bits that must be used or representing image data. Failure to respect this limits results in serious degradation o the signal to noise ratio o pyramid images. In this paper we present a theoretical analysis o the accumulated error due to repeated integer coeicient convolutions with the binomial kernel. We show that the error can be seen as a random variable and we deduce a probabilistic model that describes it. Experimental and theoretical results demonstrate that the linear complexity algorithm using integer coeicients can be made suitable or video rate computation o a hal-octave pyramid on embedded image acquisition devices. 1. Introduction Over the last years, the pyramid and Gaussian derivative eatures have emerged as a powerul source o image eatures or computer vision and pattern recognition. Gaussian derivative eatures have been used or 3D reconstruction, image stitching, object recognition [9] [1] and detection o aces [8] and other categories o image objects. Gaussian derivative eatures are the basis or SIFT descriptor [7], the histogram o gradients [4] and the second local order structure image solid [6]. Such eatures are also used in biological vision models [5]. An inconvenience o Gaussian derivative eatures is that most algorithms require relatively expensive loatingpoint computations with algorithmic complexities o N or worse. It is possible to compute such eatures with a linear complexity algorithm using only integer coeicient convolutions. However, the use o this algorithm requires careul attention to the signal noise introduced by roundo error during repeated convolutions with integer coeicient ilters. Crowley and Stern [3] have proposed a linear complexity algorithm or computing an exactly scale invariant Gaussian pyramid using a technique reerred to as Cascade convolution with resampling. This algorithm has an O(N) complexity where N is the number o pixels in the image and produces log n (N) resampled images in which each image has an identical impulse response at a logarithmically increasing set o scales. An integer coeicient version o this algorithm, the haloctave binomial pyramid was proposed by Crowley and Ri [] in 3. A pyramid computed with this algorithm can provide Gaussian derivative image eatures using only simple sums and dierences o pixels. As a result, this algorithm is suitable or calculation o Gaussian derivative eatures on embedded computing devices. However restrictions in the available memory on such devices can lead designers to implement the algorithm with as ew as 8 bits per pixel. When used with repeated convolution, this a restricted bit length introduces serious degradation in the signal to noise ratio o the resulting pyramid. Such errors can cause important loss o reliability and precision o image descriptors based on the pyramid. In this paper we present a probabilistic model or the error resulting rom the computation o the hal-octave Gaussian pyramid ixed point integer coeicients. This model describes the error generation and propagation at each pyramid level due to recursive convolutions with a binomial kernel. Our results show that the hal-octave pyramid can be accurately computed with short ixed-point numbers. This paper is organized as ollows. Section explains the hal-octave Gaussian pyramid algorithm. Our probabilistic error model is developed in section 3, ollowed experimental veriication in section 4. Section 5 concludes with discussion o the results and its application or embedded image processing.

2 Image σ = 8 σ = 4 D Binomial Convolutions σ = 4 D Binomial Convolutions σ = σ = Resampling Level 1 Pyramid Buer derivation Pyramid Derivative σ = D Binomial Convolutions l = l + 1 Resampling Level l Figure 1. The O(N) cascade convolution pyramid computation scheme. The Hal-Octave Gaussian Pyramid In this section we describe the calculation o the haloctave Gaussian pyramid and its derivatives. A linear complexity pyramid algorithm or the calculation o Gaussian derivatives has been known since the 198 s [1] (igure 1). This algorithm involves alternatively convolving with a Gaussian support, and resampling the resulting image. The result o this algorithm is a hal-octave Gaussian pyramid. An integer coeicient version o this algorithm [] has been demonstrated using repeated convolutions o the binomial kernels [1,, 1] and [1,, 1] t. Implementations that compute such pyramids on PAL sized images and bigger exist or the current generation o computer stations, and can be embedded in dedicated signal processor. The hal-octave Gaussian pyramid or an N N pixels image is composed o up to L = log (N) images. The eect o cascade convolution is to sum the variances o the ilters, so that the cumulative variance or the image l [[ 1, L ]] is σ l = l. That or, the convolution step involves convolutions by [1,, 1] in each direction. Interleaving resampling with convolutions decreases the number o image samples while expanding the distance between samples. This has the eect o dilating the Gaussian support without increasing the number o samples used or the Gaussian, eectively increasing the scale. The resampling rate is set to, resulting in a constant ratio o scale over sample distance. The resulting pyramid represents the original N N pixels image with a sequence o log (N) images at a geometric progression o scales each with twice as less samples as the previous, resulting in a total o N samples (igure ). The -dimensional derivative o order (d X, d Y ) is then obtained by convolving d X times the resulting pyramid with the derivative kernel [ 1,, 1], and d Y times with [ 1,, 1] t. Figure. The hal-octave pyramid algorithm result... xk 1 xk xk xk(i + 1, j)... xk 1 xk xk xk(i, j + 1) Figure 3. Convolution with the two base kernels (smoothing and dierentiation ). The n operator stands or right n bits shit with zero padding 3. Probabilistic Formulation 3.1. Random Error Model In this section, we present our model to analyze the noise eect o using ixed point coeicients or the pixel values in the pyramid. Let x k ( i, j) be the value o the pixel o coordinate k ater i smoothing convolutions and j dierentiations in directions X and Y. x k ( i, j) is the sum o an exact theoretical value x k ( i, j) and an error E k ( i, j). x k ( i, j) = x k ( i, j) + E k ( i, j) (1) We model the error E by a random variable. In order to simpliy the problem, the error part o contiguous pixels is supposed independent. This approximation is veriied to be very accurate in section 4. Thus, we will simply consider E where i and j are the number o smoothing and dierentiation convolutions respectively applied to one pixel, including both directions X and Y. We consider only two operations: convolution with the normalized kernels K s = [ 1 4, 1, 1 4 ] and with K d = [ 1,, 1 ] as schematized on igure 3. We also suppose that all the convolutions with K s are applied beore those with K d. Thus, we express E(i, j = ) as a unction o E(i 1, j), then E(i, j > ) as a unction o E(i, j 1).

3 3.. Error Generation and Propagation x is a ixed point number with n i bits o integer part and n bits o ractional part. Furthermore, x is supposed to be initially an integer (i.e. its n least signiicant bits are ). Because pyramid computation is recursive, the error o one value is composed o two parts. One is a new error, generated by the current computation, and the other, E prop, is accumulated during the previous computations and propagated to the new value. In this section, we determine, or a given number o ractional bits n, the expressions o the random variables En new and. Error generation is the error generated by the calculation o a new pyramid value. As shown in igure 3, a convolution with K s or K d starts with divisions by a number n shit, which, in ixed point arithmetic, corresponds to a bit shit with the loss n o the n shit least signiicant bits. Such a division can be modeled with U Vn (n shit ), the uniorm discrete random variable { } with the set o possible values V n (n shit ) = i n +n shit. In the case o values being i [, n shit 1] initially integers, this model is well-suited or i + j > n. Beore this condition is true, the divisions do not lead to a loss o inormation. Thus, the generated error is equal to: I i + j n : Otherwise: n = () { E new n (i, j = ) = U Vn () + U Vn (1) + U Vn () n (i, j > ) = U Vn (1) U Vn (1) Error propagation This paragraph considers the error that propagates rom a state to the next (i + 1, j) or (i, j + 1). We assume the number o integer bits, n i, is suicient to avoid overlow. In this case, the error propagated during a convolution with K s or K d is a random variable deined by recurrence: (3) 3.3. Total Accumulated Error The total error is the sum o the propagated part and the generated part: E n = + En new (5) We deduce rom the above model the error characteristics (standard deviation and variance) as a unction o the number o smoothing convolutions (i) and dierentiations (j). We irst express the error with recurrence ormulation, and then give the solution. Error evolution The values are initially integers, E n = or i + j n. Thus we express the expected value E[E n ] and the variance V ar[e n ] o the complete error random variable or i + j n. In this condition, and En new can be supposed independent. Smoothing convolutions are computed beore computing image derivatives rom image dierences. Thus we irst give the recurrence law or smoothing (j = ) and then or derivative computation (j > ). Smoothing: E[E n (i, j = )] = E + E En new = E [ E n (i 1, j) ] 1 n V ar[e n (i, j = )] = V ar + V ar En new Derivation: E[E n (i, j > )] = E = 3 8 V ar [ E n (i 1, j) ] + 5/3 (n +5) + E En new (6) (7) = (8) V ar[e n (i, j > )] = V ar + V ar En new = 1 V ar [ E n (i, j 1) ] + /3 (n +) (9) E prop n (, ) = (i, j = ) = En (i 1,j) 4 + En (i 1,j) + En (i 1,j) 4 (i, j > ) = En (i,j 1) En (i,j 1) (4) Error expression We can deduce the absolute expressions o E[E n ] and V ar[e n ] as a unction o i, j and n. E[E n ] = ( n i ) i j = otherwise { 1 n (1)

4 V ar[e n ] = 1 3 ) i n 1 ( 3 8 n + 1 +j+ ( 1 j ) (11) n + level 1 level 1, nd derivative Note that the error variance increases with i and j toward a limit: V ar[e n (, )] = V ar[e n (i, )] = 1/3 n + 4. Experiments and validation level level 6, nd derivative In this section we compare the error predicted by our model with the one measured by subtracting an image pyramid with varying precision ( n 1) and a 64 bits one. We perorm this analyse on images (Perlin noise) and photos randomly downloaded rom the internet Error Distribution To validate our model, we compare predicted and measured error. There is no simple mathematic expression or the distribution o the error as modeled in this paper. However we can estimate it as accurately as needed by running s o our random variable. As shown igure 4, our model accurately matches the error distribution measured in pyramids built rom highlyinormative images like Perlin noise. In the case o, an unpredicted null-error peak can be observed. The igure 5 demonstrates that in homogeneous areas (sky, unocused background, over/under-exposition,...), smoothing convolutions do not alter pixel values which thus stay integers. In such areas, our model or the loss o inormation as an addition by a random uniorm variable ails while the pyramid level is under the singular zone characteristic scale. Thus, our theoretical analysis provides inormation about the accumulated error in the worst case. One can also observe a larger spread in the measured distributions than in the theoretical one. This dierence is due the independence hypothesis which underestimate the variance by neglecting a positive covariance term. However, it appears that this term play a minor part. 4.. Error Parameters In this section, we graphically display the parameters o the random error variable. Due to the presence o homogeneous areas in photos, experimental standard deviation and variance are very datadependant. It is thus more interesting to visualize the theoretical values which correspond to the worst case (igure 6). It can be seen that the standard deviation convergence in the levels is quite ast. We thus can simply study the limit value (igure 7) Figure 4. Distribution o the error o the pyramid values with n = 6. abscissa: error value, ordinate: pixel rate with this error. (a ) (b ) Figure 5. Over-exposed photography and associated error image at pyramid level 3 with n = 6 (a ). The saturated area corresponds to a null-error area which is very poorly modeled by our random error variable. Perlin noise and its associated error image (b ). 5. Conclusion This paper has presented a model or the error due to recursive truncations by a random noise in computing a Gaussian pyramid using a linear complexity algorithm involving cascaded convolution with integer coeicient binomial ilters. This model predicts the additive image error as a unction o the number o ractional bits used, the scale o the pyramid and the order o the derivative. We have validated our model by comparing a o our random variable and measurements perormed in pyramids applied

5 Expected Value Standard Deviation n level Expected Value Standard Deviation n level Figure 6. Expected value and standard deviation o the error variable as a unction o the level and the number o ractional bits n beore dierentiations and ater dierentiations standard deviation s limit to images and photos. We have shown that the noise variance rapidly converges to a ixed value as a unction o the pyramid scale. However, the divergence is constant and grows rapidly or small numbers o ractional bits in the ixed point representation. This divergence becomes zero when the signal is dierentiated, thus making it irrelevant or many applications. We can conclude by observing that binomial convolution kernels tends to smooth the error noise, thus compensating or random noise added by integer truncation in cascade convolution. This makes the Gaussian binomial pyramid a good multi-scale image representation solution or use in embedded computer vision systems. Reerences [1] J. Crowley and A. Parker. A representation or shape based on peaks and ridges in thedierence o low-pass transorm. IEEE PAMI, 6(): , March [] J. Crowley and O. Ri. Fast computation o scale normalised gaussian receptive ields. pages , 3. 1, [3] J. Crowley and R. Stern. Fast computation o the dierence o low-pass transorm. PAMI, 6():1, March [4] N. Dalal and B. Triggs. Histograms o oriented gradients or human detection. IEEE CVPR, pages , June 5. 1 [5] M. A. Georgeson, K. A. May, T. C. Freeman, and G. S. Hesse. From ilters to eatures: scale-space analysis o edge and blur coding in human vision. Journal o vision, 7(13), 7. 1 [6] L. Griin. The second order local-image-structure solid. IEEE PAMI, 9(8): , 7. 1 [7] D. G. Lowe. Distinctive image eatures rom scale-invariant keypoints. International Journal o Computer Vision, 6:91 11, 4. 1 [8] J. A. Ruiz Hernandez, A. Lux, and J. L. Crowley. Face detection by cascade o gaussian derivatives classiiers calculated with a hal-octave pyramid. IEEE Conerence on Automatic Face and Gesture Recognition, Amsterdam, Sep 8. 1 [9] B. Schiele and J. Crowley. Recognition without correspondence using multidimensional receptive ield histograms. International Journal o Computer Vision, 36:31 5,. 1 [1] J. Yokono and T. Poggio. Oriented ilters or object recognition: an empirical study. In Proceedings o the IEEE FG4. Seoul, Korea, page 755, n Figure 7. Limit value o the error standard deviation as a unction o the number o ractional bits n.

Digital Image Processing. Lecture 6 (Enhancement) Bu-Ali Sina University Computer Engineering Dep. Fall 2009

Digital Image Processing. Lecture 6 (Enhancement) Bu-Ali Sina University Computer Engineering Dep. Fall 2009 Digital Image Processing Lecture 6 (Enhancement) Bu-Ali Sina University Computer Engineering Dep. Fall 009 Outline Image Enhancement in Spatial Domain Spatial Filtering Smoothing Filters Median Filter

More information

Lecture 12. Local Feature Detection. Matching with Invariant Features. Why extract features? Why extract features? Why extract features?

Lecture 12. Local Feature Detection. Matching with Invariant Features. Why extract features? Why extract features? Why extract features? Lecture 1 Why extract eatures? Motivation: panorama stitching We have two images how do we combine them? Local Feature Detection Guest lecturer: Alex Berg Reading: Harris and Stephens David Lowe IJCV We

More information

Estimation and detection of a periodic signal

Estimation and detection of a periodic signal Estimation and detection o a periodic signal Daniel Aronsson, Erik Björnemo, Mathias Johansson Signals and Systems Group, Uppsala University, Sweden, e-mail: Daniel.Aronsson,Erik.Bjornemo,Mathias.Johansson}@Angstrom.uu.se

More information

COMP 408/508. Computer Vision Fall 2017 PCA for Recognition

COMP 408/508. Computer Vision Fall 2017 PCA for Recognition COMP 408/508 Computer Vision Fall 07 PCA or Recognition Recall: Color Gradient by PCA v λ ( G G, ) x x x R R v, v : eigenvectors o D D with v ^v (, ) x x λ, λ : eigenvalues o D D with λ >λ v λ ( B B, )

More information

Least-Squares Spectral Analysis Theory Summary

Least-Squares Spectral Analysis Theory Summary Least-Squares Spectral Analysis Theory Summary Reerence: Mtamakaya, J. D. (2012). Assessment o Atmospheric Pressure Loading on the International GNSS REPRO1 Solutions Periodic Signatures. Ph.D. dissertation,

More information

Application of Wavelet Transform Modulus Maxima in Raman Distributed Temperature Sensors

Application of Wavelet Transform Modulus Maxima in Raman Distributed Temperature Sensors PHOTONIC SENSORS / Vol. 4, No. 2, 2014: 142 146 Application o Wavelet Transorm Modulus Maxima in Raman Distributed Temperature Sensors Zongliang WANG, Jun CHANG *, Sasa ZHANG, Sha LUO, Cuanwu JIA, Boning

More information

Lecture Outline. Basics of Spatial Filtering Smoothing Spatial Filters. Sharpening Spatial Filters

Lecture Outline. Basics of Spatial Filtering Smoothing Spatial Filters. Sharpening Spatial Filters 1 Lecture Outline Basics o Spatial Filtering Smoothing Spatial Filters Averaging ilters Order-Statistics ilters Sharpening Spatial Filters Laplacian ilters High-boost ilters Gradient Masks Combining Spatial

More information

Enhancement Using Local Histogram

Enhancement Using Local Histogram Enhancement Using Local Histogram Used to enhance details over small portions o the image. Deine a square or rectangular neighborhood hose center moves rom piel to piel. Compute local histogram based on

More information

Bayesian Technique for Reducing Uncertainty in Fatigue Failure Model

Bayesian Technique for Reducing Uncertainty in Fatigue Failure Model 9IDM- Bayesian Technique or Reducing Uncertainty in Fatigue Failure Model Sriram Pattabhiraman and Nam H. Kim University o Florida, Gainesville, FL, 36 Copyright 8 SAE International ABSTRACT In this paper,

More information

TLT-5200/5206 COMMUNICATION THEORY, Exercise 3, Fall TLT-5200/5206 COMMUNICATION THEORY, Exercise 3, Fall Problem 1.

TLT-5200/5206 COMMUNICATION THEORY, Exercise 3, Fall TLT-5200/5206 COMMUNICATION THEORY, Exercise 3, Fall Problem 1. TLT-5/56 COMMUNICATION THEORY, Exercise 3, Fall Problem. The "random walk" was modelled as a random sequence [ n] where W[i] are binary i.i.d. random variables with P[W[i] = s] = p (orward step with probability

More information

Midterm Summary Fall 08. Yao Wang Polytechnic University, Brooklyn, NY 11201

Midterm Summary Fall 08. Yao Wang Polytechnic University, Brooklyn, NY 11201 Midterm Summary Fall 8 Yao Wang Polytechnic University, Brooklyn, NY 2 Components in Digital Image Processing Output are images Input Image Color Color image image processing Image Image restoration Image

More information

Ultra Fast Calculation of Temperature Profiles of VLSI ICs in Thermal Packages Considering Parameter Variations

Ultra Fast Calculation of Temperature Profiles of VLSI ICs in Thermal Packages Considering Parameter Variations Ultra Fast Calculation o Temperature Proiles o VLSI ICs in Thermal Packages Considering Parameter Variations Je-Hyoung Park, Virginia Martín Hériz, Ali Shakouri, and Sung-Mo Kang Dept. o Electrical Engineering,

More information

Numerical Methods - Lecture 2. Numerical Methods. Lecture 2. Analysis of errors in numerical methods

Numerical Methods - Lecture 2. Numerical Methods. Lecture 2. Analysis of errors in numerical methods Numerical Methods - Lecture 1 Numerical Methods Lecture. Analysis o errors in numerical methods Numerical Methods - Lecture Why represent numbers in loating point ormat? Eample 1. How a number 56.78 can

More information

CHAPTER 1: INTRODUCTION. 1.1 Inverse Theory: What It Is and What It Does

CHAPTER 1: INTRODUCTION. 1.1 Inverse Theory: What It Is and What It Does Geosciences 567: CHAPTER (RR/GZ) CHAPTER : INTRODUCTION Inverse Theory: What It Is and What It Does Inverse theory, at least as I choose to deine it, is the ine art o estimating model parameters rom data

More information

Additional exercises in Stationary Stochastic Processes

Additional exercises in Stationary Stochastic Processes Mathematical Statistics, Centre or Mathematical Sciences Lund University Additional exercises 8 * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *

More information

A Comparative Study of Non-separable Wavelet and Tensor-product. Wavelet; Image Compression

A Comparative Study of Non-separable Wavelet and Tensor-product. Wavelet; Image Compression Copyright c 007 Tech Science Press CMES, vol., no., pp.91-96, 007 A Comparative Study o Non-separable Wavelet and Tensor-product Wavelet in Image Compression Jun Zhang 1 Abstract: The most commonly used

More information

LoG Blob Finding and Scale. Scale Selection. Blobs (and scale selection) Achieving scale covariance. Blob detection in 2D. Blob detection in 2D

LoG Blob Finding and Scale. Scale Selection. Blobs (and scale selection) Achieving scale covariance. Blob detection in 2D. Blob detection in 2D Achieving scale covariance Blobs (and scale selection) Goal: independently detect corresponding regions in scaled versions of the same image Need scale selection mechanism for finding characteristic region

More information

Invariant local features. Invariant Local Features. Classes of transformations. (Good) invariant local features. Case study: panorama stitching

Invariant local features. Invariant Local Features. Classes of transformations. (Good) invariant local features. Case study: panorama stitching Invariant local eatures Invariant Local Features Tuesday, February 6 Subset o local eature types designed to be invariant to Scale Translation Rotation Aine transormations Illumination 1) Detect distinctive

More information

The Deutsch-Jozsa Problem: De-quantization and entanglement

The Deutsch-Jozsa Problem: De-quantization and entanglement The Deutsch-Jozsa Problem: De-quantization and entanglement Alastair A. Abbott Department o Computer Science University o Auckland, New Zealand May 31, 009 Abstract The Deustch-Jozsa problem is one o the

More information

Curve Sketching. The process of curve sketching can be performed in the following steps:

Curve Sketching. The process of curve sketching can be performed in the following steps: Curve Sketching So ar you have learned how to ind st and nd derivatives o unctions and use these derivatives to determine where a unction is:. Increasing/decreasing. Relative extrema 3. Concavity 4. Points

More information

3. Several Random Variables

3. Several Random Variables . Several Random Variables. Two Random Variables. Conditional Probabilit--Revisited. Statistical Independence.4 Correlation between Random Variables. Densit unction o the Sum o Two Random Variables. Probabilit

More information

CHAPTER 8 ANALYSIS OF AVERAGE SQUARED DIFFERENCE SURFACES

CHAPTER 8 ANALYSIS OF AVERAGE SQUARED DIFFERENCE SURFACES CAPTER 8 ANALYSS O AVERAGE SQUARED DERENCE SURACES n Chapters 5, 6, and 7, the Spectral it algorithm was used to estimate both scatterer size and total attenuation rom the backscattered waveorms by minimizing

More information

Achieving scale covariance

Achieving scale covariance Achieving scale covariance Goal: independently detect corresponding regions in scaled versions of the same image Need scale selection mechanism for finding characteristic region size that is covariant

More information

In many diverse fields physical data is collected or analysed as Fourier components.

In many diverse fields physical data is collected or analysed as Fourier components. 1. Fourier Methods In many diverse ields physical data is collected or analysed as Fourier components. In this section we briely discuss the mathematics o Fourier series and Fourier transorms. 1. Fourier

More information

Scattered Data Approximation of Noisy Data via Iterated Moving Least Squares

Scattered Data Approximation of Noisy Data via Iterated Moving Least Squares Scattered Data Approximation o Noisy Data via Iterated Moving Least Squares Gregory E. Fasshauer and Jack G. Zhang Abstract. In this paper we ocus on two methods or multivariate approximation problems

More information

Analysis of Integer Transformation and Quantization Blocks using H.264 Standard and the Conventional DCT Techniques

Analysis of Integer Transformation and Quantization Blocks using H.264 Standard and the Conventional DCT Techniques Priyanka P James et al, International Journal o Computer Science and Mobile Computing, Vol.3 Issue.3, March- 2014, pg. 873-878 Available Online at www.ijcsmc.com International Journal o Computer Science

More information

Image Enhancement (Spatial Filtering 2)

Image Enhancement (Spatial Filtering 2) Image Enhancement (Spatial Filtering ) Dr. Samir H. Abdul-Jauwad Electrical Engineering Department College o Engineering Sciences King Fahd University o Petroleum & Minerals Dhahran Saudi Arabia samara@kupm.edu.sa

More information

Aggregate Growth: R =αn 1/ d f

Aggregate Growth: R =αn 1/ d f Aggregate Growth: Mass-ractal aggregates are partly described by the mass-ractal dimension, d, that deines the relationship between size and mass, R =αn 1/ d where α is the lacunarity constant, R is the

More information

Kalman filtering based probabilistic nowcasting of object oriented tracked convective storms

Kalman filtering based probabilistic nowcasting of object oriented tracked convective storms Kalman iltering based probabilistic nowcasting o object oriented traced convective storms Pea J. Rossi,3, V. Chandrasear,2, Vesa Hasu 3 Finnish Meteorological Institute, Finland, Eri Palménin Auio, pea.rossi@mi.i

More information

IMPROVED NOISE CANCELLATION IN DISCRETE COSINE TRANSFORM DOMAIN USING ADAPTIVE BLOCK LMS FILTER

IMPROVED NOISE CANCELLATION IN DISCRETE COSINE TRANSFORM DOMAIN USING ADAPTIVE BLOCK LMS FILTER SANJAY KUMAR GUPTA* et al. ISSN: 50 3676 [IJESAT] INTERNATIONAL JOURNAL OF ENGINEERING SCIENCE & ADVANCED TECHNOLOGY Volume-, Issue-3, 498 50 IMPROVED NOISE CANCELLATION IN DISCRETE COSINE TRANSFORM DOMAIN

More information

New method for two-point nonuniformity correction of microbolometer detectors

New method for two-point nonuniformity correction of microbolometer detectors 10 th International Conerence on Quantitative InraRed Thermography July 27-30, 2010, Québec (Canada) New method or two-point nonuniormity correction o microbolometer detectors by R. Olbrycht*, B. Wiecek*,

More information

APPENDIX 1 ERROR ESTIMATION

APPENDIX 1 ERROR ESTIMATION 1 APPENDIX 1 ERROR ESTIMATION Measurements are always subject to some uncertainties no matter how modern and expensive equipment is used or how careully the measurements are perormed These uncertainties

More information

arxiv: v1 [cs.it] 12 Mar 2014

arxiv: v1 [cs.it] 12 Mar 2014 COMPRESSIVE SIGNAL PROCESSING WITH CIRCULANT SENSING MATRICES Diego Valsesia Enrico Magli Politecnico di Torino (Italy) Dipartimento di Elettronica e Telecomunicazioni arxiv:403.2835v [cs.it] 2 Mar 204

More information

Syllabus Objective: 2.9 The student will sketch the graph of a polynomial, radical, or rational function.

Syllabus Objective: 2.9 The student will sketch the graph of a polynomial, radical, or rational function. Precalculus Notes: Unit Polynomial Functions Syllabus Objective:.9 The student will sketch the graph o a polynomial, radical, or rational unction. Polynomial Function: a unction that can be written in

More information

BANDELET IMAGE APPROXIMATION AND COMPRESSION

BANDELET IMAGE APPROXIMATION AND COMPRESSION BANDELET IMAGE APPOXIMATION AND COMPESSION E. LE PENNEC AND S. MALLAT Abstract. Finding eicient geometric representations o images is a central issue to improve image compression and noise removal algorithms.

More information

Intrinsic Small-Signal Equivalent Circuit of GaAs MESFET s

Intrinsic Small-Signal Equivalent Circuit of GaAs MESFET s Intrinsic Small-Signal Equivalent Circuit o GaAs MESFET s M KAMECHE *, M FEHAM M MELIANI, N BENAHMED, S DALI * National Centre o Space Techniques, Algeria Telecom Laboratory, University o Tlemcen, Algeria

More information

On High-Rate Cryptographic Compression Functions

On High-Rate Cryptographic Compression Functions On High-Rate Cryptographic Compression Functions Richard Ostertág and Martin Stanek Department o Computer Science Faculty o Mathematics, Physics and Inormatics Comenius University Mlynská dolina, 842 48

More information

Two-step self-tuning phase-shifting interferometry

Two-step self-tuning phase-shifting interferometry Two-step sel-tuning phase-shiting intererometry J. Vargas, 1,* J. Antonio Quiroga, T. Belenguer, 1 M. Servín, 3 J. C. Estrada 3 1 Laboratorio de Instrumentación Espacial, Instituto Nacional de Técnica

More information

Introduction to Analog And Digital Communications

Introduction to Analog And Digital Communications Introduction to Analog And Digital Communications Second Edition Simon Haykin, Michael Moher Chapter Fourier Representation o Signals and Systems.1 The Fourier Transorm. Properties o the Fourier Transorm.3

More information

Lecture 13: Applications of Fourier transforms (Recipes, Chapter 13)

Lecture 13: Applications of Fourier transforms (Recipes, Chapter 13) Lecture 13: Applications o Fourier transorms (Recipes, Chapter 13 There are many applications o FT, some o which involve the convolution theorem (Recipes 13.1: The convolution o h(t and r(t is deined by

More information

Objectives. By the time the student is finished with this section of the workbook, he/she should be able

Objectives. By the time the student is finished with this section of the workbook, he/she should be able FUNCTIONS Quadratic Functions......8 Absolute Value Functions.....48 Translations o Functions..57 Radical Functions...61 Eponential Functions...7 Logarithmic Functions......8 Cubic Functions......91 Piece-Wise

More information

Local Heat Transfer Coefficient Measurements, Using a Transient Imaging Method With an Inverse Scheme

Local Heat Transfer Coefficient Measurements, Using a Transient Imaging Method With an Inverse Scheme Local Heat Transer Coeicient Measurements, Using a Transient Imaging Method With an Inverse Scheme A. EL ABBADI, D. BOUGEARD, B.BAUDOIN Ecole des Mines de Douai, Département Energétique Industrielle, 941,

More information

ELEG 3143 Probability & Stochastic Process Ch. 4 Multiple Random Variables

ELEG 3143 Probability & Stochastic Process Ch. 4 Multiple Random Variables Department o Electrical Engineering University o Arkansas ELEG 3143 Probability & Stochastic Process Ch. 4 Multiple Random Variables Dr. Jingxian Wu wuj@uark.edu OUTLINE 2 Two discrete random variables

More information

8. INTRODUCTION TO STATISTICAL THERMODYNAMICS

8. INTRODUCTION TO STATISTICAL THERMODYNAMICS n * D n d Fluid z z z FIGURE 8-1. A SYSTEM IS IN EQUILIBRIUM EVEN IF THERE ARE VARIATIONS IN THE NUMBER OF MOLECULES IN A SMALL VOLUME, SO LONG AS THE PROPERTIES ARE UNIFORM ON A MACROSCOPIC SCALE 8. INTRODUCTION

More information

Exponential and Logarithmic. Functions CHAPTER The Algebra of Functions; Composite

Exponential and Logarithmic. Functions CHAPTER The Algebra of Functions; Composite CHAPTER 9 Exponential and Logarithmic Functions 9. The Algebra o Functions; Composite Functions 9.2 Inverse Functions 9.3 Exponential Functions 9.4 Exponential Growth and Decay Functions 9.5 Logarithmic

More information

Random Error Analysis of Inertial Sensors output Based on Allan Variance Shaochen Li1, a, Xiaojing Du2,b and Junyi Zhai3,c

Random Error Analysis of Inertial Sensors output Based on Allan Variance Shaochen Li1, a, Xiaojing Du2,b and Junyi Zhai3,c International Conerence on Civil, Transportation and Environment (ICCTE 06) Random Error Analysis o Inertial Sensors output Based on Allan Variance Shaochen Li, a, Xiaojing Du, and Junyi Zhai3,c School

More information

MODULE 6 LECTURE NOTES 1 REVIEW OF PROBABILITY THEORY. Most water resources decision problems face the risk of uncertainty mainly because of the

MODULE 6 LECTURE NOTES 1 REVIEW OF PROBABILITY THEORY. Most water resources decision problems face the risk of uncertainty mainly because of the MODULE 6 LECTURE NOTES REVIEW OF PROBABILITY THEORY INTRODUCTION Most water resources decision problems ace the risk o uncertainty mainly because o the randomness o the variables that inluence the perormance

More information

Estimation of Human Emotions Using Thermal Facial Information

Estimation of Human Emotions Using Thermal Facial Information Estimation o Human Emotions Using Thermal acial Inormation Hung Nguyen 1, Kazunori Kotani 1, an Chen 1, Bac Le 2 1 Japan Advanced Institute o Science and Technology, 1-1 Asahidai, Nomi, Ishikawa, Japan

More information

SIFT: SCALE INVARIANT FEATURE TRANSFORM BY DAVID LOWE

SIFT: SCALE INVARIANT FEATURE TRANSFORM BY DAVID LOWE SIFT: SCALE INVARIANT FEATURE TRANSFORM BY DAVID LOWE Overview Motivation of Work Overview of Algorithm Scale Space and Difference of Gaussian Keypoint Localization Orientation Assignment Descriptor Building

More information

Supplementary material for Continuous-action planning for discounted infinite-horizon nonlinear optimal control with Lipschitz values

Supplementary material for Continuous-action planning for discounted infinite-horizon nonlinear optimal control with Lipschitz values Supplementary material or Continuous-action planning or discounted ininite-horizon nonlinear optimal control with Lipschitz values List o main notations x, X, u, U state, state space, action, action space,

More information

Reliability Assessment with Correlated Variables using Support Vector Machines

Reliability Assessment with Correlated Variables using Support Vector Machines Reliability Assessment with Correlated Variables using Support Vector Machines Peng Jiang, Anirban Basudhar, and Samy Missoum Aerospace and Mechanical Engineering Department, University o Arizona, Tucson,

More information

NONPARAMETRIC PREDICTIVE INFERENCE FOR REPRODUCIBILITY OF TWO BASIC TESTS BASED ON ORDER STATISTICS

NONPARAMETRIC PREDICTIVE INFERENCE FOR REPRODUCIBILITY OF TWO BASIC TESTS BASED ON ORDER STATISTICS REVSTAT Statistical Journal Volume 16, Number 2, April 2018, 167 185 NONPARAMETRIC PREDICTIVE INFERENCE FOR REPRODUCIBILITY OF TWO BASIC TESTS BASED ON ORDER STATISTICS Authors: Frank P.A. Coolen Department

More information

RESOLUTION MSC.362(92) (Adopted on 14 June 2013) REVISED RECOMMENDATION ON A STANDARD METHOD FOR EVALUATING CROSS-FLOODING ARRANGEMENTS

RESOLUTION MSC.362(92) (Adopted on 14 June 2013) REVISED RECOMMENDATION ON A STANDARD METHOD FOR EVALUATING CROSS-FLOODING ARRANGEMENTS (Adopted on 4 June 203) (Adopted on 4 June 203) ANNEX 8 (Adopted on 4 June 203) MSC 92/26/Add. Annex 8, page THE MARITIME SAFETY COMMITTEE, RECALLING Article 28(b) o the Convention on the International

More information

Edges and Scale. Image Features. Detecting edges. Origin of Edges. Solution: smooth first. Effects of noise

Edges and Scale. Image Features. Detecting edges. Origin of Edges. Solution: smooth first. Effects of noise Edges and Scale Image Features From Sandlot Science Slides revised from S. Seitz, R. Szeliski, S. Lazebnik, etc. Origin of Edges surface normal discontinuity depth discontinuity surface color discontinuity

More information

Estimation of Sample Reactivity Worth with Differential Operator Sampling Method

Estimation of Sample Reactivity Worth with Differential Operator Sampling Method Progress in NUCLEAR SCIENCE and TECHNOLOGY, Vol. 2, pp.842-850 (2011) ARTICLE Estimation o Sample Reactivity Worth with Dierential Operator Sampling Method Yasunobu NAGAYA and Takamasa MORI Japan Atomic

More information

RELIABILITY OF BURIED PIPELINES WITH CORROSION DEFECTS UNDER VARYING BOUNDARY CONDITIONS

RELIABILITY OF BURIED PIPELINES WITH CORROSION DEFECTS UNDER VARYING BOUNDARY CONDITIONS REIABIITY OF BURIE PIPEIES WITH CORROSIO EFECTS UER VARYIG BOUARY COITIOS Ouk-Sub ee 1 and ong-hyeok Kim 1. School o Mechanical Engineering, InHa University #53, Yonghyun-ong, am-ku, Incheon, 40-751, Korea

More information

Available online at ScienceDirect. Procedia Engineering 105 (2015 )

Available online at  ScienceDirect. Procedia Engineering 105 (2015 ) Available online at www.sciencedirect.com ScienceDirect Procedia Engineering 105 (2015 ) 388 397 6th BSME International Conerence on Thermal Engineering (ICTE 2014) Eect o tilt angle on pure mixed convection

More information

Analysis Scheme in the Ensemble Kalman Filter

Analysis Scheme in the Ensemble Kalman Filter JUNE 1998 BURGERS ET AL. 1719 Analysis Scheme in the Ensemble Kalman Filter GERRIT BURGERS Royal Netherlands Meteorological Institute, De Bilt, the Netherlands PETER JAN VAN LEEUWEN Institute or Marine

More information

Chapter 3: Image Enhancement in the. Office room : 841

Chapter 3: Image Enhancement in the.   Office room : 841 Chapter 3: Image Enhancement in the Spatial Domain Lecturer: Jianbing Shen Email : shenjianbing@bit.edu.cn Oice room : 841 http://cs.bit.edu.cn/shenjianbing cn/shenjianbing Principle Objective o Enhancement

More information

Gaussian Process Regression Models for Predicting Stock Trends

Gaussian Process Regression Models for Predicting Stock Trends Gaussian Process Regression Models or Predicting Stock Trends M. Todd Farrell Andrew Correa December 5, 7 Introduction Historical stock price data is a massive amount o time-series data with little-to-no

More information

Binary Pressure-Sensitive Paint

Binary Pressure-Sensitive Paint Binary Pressure-ensitive Paint It is well documented in the literature that pressure sensitive paints exhibit undesirable sensitivity to variations in temperature and illumination. In act these variations

More information

Robust Residual Selection for Fault Detection

Robust Residual Selection for Fault Detection Robust Residual Selection or Fault Detection Hamed Khorasgani*, Daniel E Jung**, Gautam Biswas*, Erik Frisk**, and Mattias Krysander** Abstract A number o residual generation methods have been developed

More information

Notes on Wavelets- Sandra Chapman (MPAGS: Time series analysis) # $ ( ) = G f. y t

Notes on Wavelets- Sandra Chapman (MPAGS: Time series analysis) # $ ( ) = G f. y t Wavelets Recall: we can choose! t ) as basis on which we expand, ie: ) = y t ) = G! t ) y t! may be orthogonal chosen or appropriate properties. This is equivalent to the transorm: ) = G y t )!,t )d 2

More information

System Identification & Parameter Estimation

System Identification & Parameter Estimation System Identiication & Parameter Estimation Wb30: SIPE Lecture 9: Physical Modeling, Model and Parameter Accuracy Erwin de Vlugt, Dept. o Biomechanical Engineering BMechE, Fac. 3mE April 6 00 Delt University

More information

Physics 5153 Classical Mechanics. Solution by Quadrature-1

Physics 5153 Classical Mechanics. Solution by Quadrature-1 October 14, 003 11:47:49 1 Introduction Physics 5153 Classical Mechanics Solution by Quadrature In the previous lectures, we have reduced the number o eective degrees o reedom that are needed to solve

More information

Vlad Estivill-Castro (2016) Robots for People --- A project for intelligent integrated systems

Vlad Estivill-Castro (2016) Robots for People --- A project for intelligent integrated systems 1 Vlad Estivill-Castro (2016) Robots for People --- A project for intelligent integrated systems V. Estivill-Castro 2 Perception Concepts Vision Chapter 4 (textbook) Sections 4.3 to 4.5 What is the course

More information

ITERATED SRINKAGE ALGORITHM FOR BASIS PURSUIT MINIMIZATION

ITERATED SRINKAGE ALGORITHM FOR BASIS PURSUIT MINIMIZATION ITERATED SRINKAGE ALGORITHM FOR BASIS PURSUIT MINIMIZATION Michael Elad The Computer Science Department The Technion Israel Institute o technology Haia 3000, Israel * SIAM Conerence on Imaging Science

More information

( 1) ( 2) ( 1) nan integer, since the potential is no longer simple harmonic.

( 1) ( 2) ( 1) nan integer, since the potential is no longer simple harmonic. . Anharmonic Oscillators Michael Fowler Landau (para 8) considers a simple harmonic oscillator with added small potential energy terms mα + mβ. We ll simpliy slightly by dropping the term, to give an equation

More information

8.4 Inverse Functions

8.4 Inverse Functions Section 8. Inverse Functions 803 8. Inverse Functions As we saw in the last section, in order to solve application problems involving eponential unctions, we will need to be able to solve eponential equations

More information

STAT 801: Mathematical Statistics. Hypothesis Testing

STAT 801: Mathematical Statistics. Hypothesis Testing STAT 801: Mathematical Statistics Hypothesis Testing Hypothesis testing: a statistical problem where you must choose, on the basis o data X, between two alternatives. We ormalize this as the problem o

More information

AP Calculus Notes: Unit 1 Limits & Continuity. Syllabus Objective: 1.1 The student will calculate limits using the basic limit theorems.

AP Calculus Notes: Unit 1 Limits & Continuity. Syllabus Objective: 1.1 The student will calculate limits using the basic limit theorems. Syllabus Objective:. The student will calculate its using the basic it theorems. LIMITS how the outputs o a unction behave as the inputs approach some value Finding a Limit Notation: The it as approaches

More information

A new Control Strategy for Trajectory Tracking of Fire Rescue Turntable Ladders

A new Control Strategy for Trajectory Tracking of Fire Rescue Turntable Ladders Proceedings o the 17th World Congress The International Federation o Automatic Control Seoul, Korea, July 6-11, 28 A new Control Strategy or Trajectory Tracking o Fire Rescue Turntable Ladders N. Zimmert,

More information

Feedback linearization control of systems with singularities: a ball-beam revisit

Feedback linearization control of systems with singularities: a ball-beam revisit Chapter 1 Feedback linearization control o systems with singularities: a ball-beam revisit Fu Zhang The Mathworks, Inc zhang@mathworks.com Benito Fernndez-Rodriguez Department o Mechanical Engineering,

More information

AH 2700A. Attenuator Pair Ratio for C vs Frequency. Option-E 50 Hz-20 khz Ultra-precision Capacitance/Loss Bridge

AH 2700A. Attenuator Pair Ratio for C vs Frequency. Option-E 50 Hz-20 khz Ultra-precision Capacitance/Loss Bridge 0 E ttenuator Pair Ratio or vs requency NEEN-ERLN 700 Option-E 0-0 k Ultra-precision apacitance/loss ridge ttenuator Ratio Pair Uncertainty o in ppm or ll Usable Pairs o Taps 0 0 0. 0. 0. 07/08/0 E E E

More information

Classification of high dimensional data: High Dimensional Discriminant Analysis

Classification of high dimensional data: High Dimensional Discriminant Analysis Classification of high dimensional data: High Dimensional Discriminant Analysis Charles Bouveyron, Stephane Girard, Cordelia Schmid To cite this version: Charles Bouveyron, Stephane Girard, Cordelia Schmid.

More information

The achievable limits of operational modal analysis. * Siu-Kui Au 1)

The achievable limits of operational modal analysis. * Siu-Kui Au 1) The achievable limits o operational modal analysis * Siu-Kui Au 1) 1) Center or Engineering Dynamics and Institute or Risk and Uncertainty, University o Liverpool, Liverpool L69 3GH, United Kingdom 1)

More information

Local Features (contd.)

Local Features (contd.) Motivation Local Features (contd.) Readings: Mikolajczyk and Schmid; F&P Ch 10 Feature points are used also or: Image alignment (homography, undamental matrix) 3D reconstruction Motion tracking Object

More information

NONLINEAR CONTROL OF POWER NETWORK MODELS USING FEEDBACK LINEARIZATION

NONLINEAR CONTROL OF POWER NETWORK MODELS USING FEEDBACK LINEARIZATION NONLINEAR CONTROL OF POWER NETWORK MODELS USING FEEDBACK LINEARIZATION Steven Ball Science Applications International Corporation Columbia, MD email: sball@nmtedu Steve Schaer Department o Mathematics

More information

Sparsity-Aware Pseudo Affine Projection Algorithm for Active Noise Control

Sparsity-Aware Pseudo Affine Projection Algorithm for Active Noise Control Sparsity-Aware Pseudo Aine Projection Algorithm or Active Noise Control Felix Albu *, Amelia Gully and Rodrigo de Lamare * Department o Electronics, Valahia University o argoviste, argoviste, Romania E-mail:

More information

Chapter 6 Reliability-based design and code developments

Chapter 6 Reliability-based design and code developments Chapter 6 Reliability-based design and code developments 6. General Reliability technology has become a powerul tool or the design engineer and is widely employed in practice. Structural reliability analysis

More information

High Dimensional Discriminant Analysis

High Dimensional Discriminant Analysis High Dimensional Discriminant Analysis Charles Bouveyron 1,2, Stéphane Girard 1, and Cordelia Schmid 2 1 LMC IMAG, BP 53, Université Grenoble 1, 38041 Grenoble cedex 9 France (e-mail: charles.bouveyron@imag.fr,

More information

Evolving Dynamic Multi-objective Optimization Problems. with Objective Replacement. Abstract

Evolving Dynamic Multi-objective Optimization Problems. with Objective Replacement. Abstract Evolving Dynamic Multi-objective Optimization Problems with Objective eplacement Sheng-Uei Guan, Qian Chen and Wenting Mo Department o Electrical and Computer Engineering National University o Singapore

More information

Symbolic-Numeric Methods for Improving Structural Analysis of DAEs

Symbolic-Numeric Methods for Improving Structural Analysis of DAEs Symbolic-Numeric Methods or Improving Structural Analysis o DAEs Guangning Tan, Nedialko S. Nedialkov, and John D. Pryce Abstract Systems o dierential-algebraic equations (DAEs) are generated routinely

More information

Mixed Signal IC Design Notes set 6: Mathematics of Electrical Noise

Mixed Signal IC Design Notes set 6: Mathematics of Electrical Noise ECE45C /8C notes, M. odwell, copyrighted 007 Mied Signal IC Design Notes set 6: Mathematics o Electrical Noise Mark odwell University o Caliornia, Santa Barbara rodwell@ece.ucsb.edu 805-893-344, 805-893-36

More information

Second Order Slip Flow of Cu-Water Nanofluid Over a Stretching Sheet With Heat Transfer

Second Order Slip Flow of Cu-Water Nanofluid Over a Stretching Sheet With Heat Transfer Second Order Slip Flow o Cu-Water Nanoluid Over a Stretching Sheet With Heat Transer RAJESH SHARMA AND ANUAR ISHAK School o Mathematical Sciences, Faculty o Science and Technology Universiti Kebangsaan

More information

Detectors part II Descriptors

Detectors part II Descriptors EECS 442 Computer vision Detectors part II Descriptors Blob detectors Invariance Descriptors Some slides of this lectures are courtesy of prof F. Li, prof S. Lazebnik, and various other lecturers Goal:

More information

Finite Dimensional Hilbert Spaces are Complete for Dagger Compact Closed Categories (Extended Abstract)

Finite Dimensional Hilbert Spaces are Complete for Dagger Compact Closed Categories (Extended Abstract) Electronic Notes in Theoretical Computer Science 270 (1) (2011) 113 119 www.elsevier.com/locate/entcs Finite Dimensional Hilbert Spaces are Complete or Dagger Compact Closed Categories (Extended bstract)

More information

New Results on Boomerang and Rectangle Attacks

New Results on Boomerang and Rectangle Attacks New Results on Boomerang and Rectangle Attacks Eli Biham, 1 Orr Dunkelman, 1 Nathan Keller 2 1 Computer Science Department, Technion. Haia 32000, Israel {biham,orrd}@cs.technion.ac.il 2 Mathematics Department,

More information

Feasibility of a Multi-Pass Thomson Scattering System with Confocal Spherical Mirrors

Feasibility of a Multi-Pass Thomson Scattering System with Confocal Spherical Mirrors Plasma and Fusion Research: Letters Volume 5, 044 200) Feasibility o a Multi-Pass Thomson Scattering System with Conocal Spherical Mirrors Junichi HIRATSUKA, Akira EJIRI, Yuichi TAKASE and Takashi YAMAGUCHI

More information

Probabilistic Engineering Mechanics

Probabilistic Engineering Mechanics Probabilistic Engineering Mechanics 6 (11) 174 181 Contents lists available at ScienceDirect Probabilistic Engineering Mechanics journal homepage: www.elsevier.com/locate/probengmech Variability response

More information

Chapter 4 Imaging. Lecture 21. d (110) Chem 793, Fall 2011, L. Ma

Chapter 4 Imaging. Lecture 21. d (110) Chem 793, Fall 2011, L. Ma Chapter 4 Imaging Lecture 21 d (110) Imaging Imaging in the TEM Diraction Contrast in TEM Image HRTEM (High Resolution Transmission Electron Microscopy) Imaging or phase contrast imaging STEM imaging a

More information

Conference Article. Spectral Properties of Chaotic Signals Generated by the Bernoulli Map

Conference Article. Spectral Properties of Chaotic Signals Generated by the Bernoulli Map Jestr Journal o Engineering Science and Technology Review 8 () (05) -6 Special Issue on Synchronization and Control o Chaos: Theory, Methods and Applications Conerence Article JOURNAL OF Engineering Science

More information

Math-3 Lesson 1-4. Review: Cube, Cube Root, and Exponential Functions

Math-3 Lesson 1-4. Review: Cube, Cube Root, and Exponential Functions Math- Lesson -4 Review: Cube, Cube Root, and Eponential Functions Quiz - Graph (no calculator):. y. y ( ) 4. y What is a power? vocabulary Power: An epression ormed by repeated Multiplication o the same

More information

A UNIFIED FRAMEWORK FOR MULTICHANNEL FAST QRD-LS ADAPTIVE FILTERS BASED ON BACKWARD PREDICTION ERRORS

A UNIFIED FRAMEWORK FOR MULTICHANNEL FAST QRD-LS ADAPTIVE FILTERS BASED ON BACKWARD PREDICTION ERRORS A UNIFIED FRAMEWORK FOR MULTICHANNEL FAST QRD-LS ADAPTIVE FILTERS BASED ON BACKWARD PREDICTION ERRORS César A Medina S,José A Apolinário Jr y, and Marcio G Siqueira IME Department o Electrical Engineering

More information

Products and Convolutions of Gaussian Probability Density Functions

Products and Convolutions of Gaussian Probability Density Functions Tina Memo No. 003-003 Internal Report Products and Convolutions o Gaussian Probability Density Functions P.A. Bromiley Last updated / 9 / 03 Imaging Science and Biomedical Engineering Division, Medical

More information

RATIONAL FUNCTIONS. Finding Asymptotes..347 The Domain Finding Intercepts Graphing Rational Functions

RATIONAL FUNCTIONS. Finding Asymptotes..347 The Domain Finding Intercepts Graphing Rational Functions RATIONAL FUNCTIONS Finding Asymptotes..347 The Domain....350 Finding Intercepts.....35 Graphing Rational Functions... 35 345 Objectives The ollowing is a list o objectives or this section o the workbook.

More information

Presenters. Time Domain and Statistical Model Development, Simulation and Correlation Methods for High Speed SerDes

Presenters. Time Domain and Statistical Model Development, Simulation and Correlation Methods for High Speed SerDes JANUARY 28-31, 2013 SANTA CLARA CONVENTION CENTER Time Domain and Statistical Model Development, Simulation and Correlation Methods or High Speed SerDes Presenters Xingdong Dai Fangyi Rao Shiva Prasad

More information

DATA ASSIMILATION IN A COMBINED 1D-2D FLOOD MODEL

DATA ASSIMILATION IN A COMBINED 1D-2D FLOOD MODEL Proceedings o the International Conerence Innovation, Advances and Implementation o Flood Forecasting Technology, Tromsø, DATA ASSIMILATION IN A COMBINED 1D-2D FLOOD MODEL Johan Hartnac, Henri Madsen and

More information

1. Cyclone III Device Family Overview

1. Cyclone III Device Family Overview 1. Cyclone III Device Family Overview CIII51001-1.3 Lowest System-Cost FPGAs The Cyclone III device amily oered by Altera is a cost-optimized, memory-rich FPGA amily. Cyclone III FPGAs are built on Taiwan

More information

An Alternative Poincaré Section for Steady-State Responses and Bifurcations of a Duffing-Van der Pol Oscillator

An Alternative Poincaré Section for Steady-State Responses and Bifurcations of a Duffing-Van der Pol Oscillator An Alternative Poincaré Section or Steady-State Responses and Biurcations o a Duing-Van der Pol Oscillator Jang-Der Jeng, Yuan Kang *, Yeon-Pun Chang Department o Mechanical Engineering, National United

More information