Vector-attribute filters

Size: px
Start display at page:

Download "Vector-attribute filters"

Transcription

1 Vector-attribute filters Erik R. Urbach, Niek J. Boersma, and Michael H.F. Wilkinson Institute for Mathematics and Computing Science University of Groningen The Netherlands April 2005

2 Outline Purpose Binary vector-attribute filters Gray-scale vector-attribute filters Moment invariants as vector-attribute Image analysis using pattern spectra Conclusions

3 Purpose Attribute filters use a criterion to remove or preserve connected components (flat zones) based on their attributes. So far attribute filters have always been based on one or more scalar attributes, e.g. area, perimeter, elongation, number of holes. Binary image X Nuts Bolts PRO : Computationally efficient for image filtering and analysis. CON : How to remove/preserve components based on their similarity with a given shape.

4 Vector-attribute filters Removing objects that are similar enough to a given shape. Example: removing objects that are similar enough (ɛ) to the reference shape (letter A). Original image X ɛ = 0.01 ɛ = 0.10 ɛ = 0.15 A value of ɛ = 0 means only those shapes are removed that are exactly the same as the reference shape.

5 Binary attribute thinning Trivial thinning Φ T of a connected set C with criterion T is C if C satisfies T, and is empty otherwise. Furthermore, Φ T ( ) =. Binary connected opening Γ x (X) of set X at point x M yields the connected component of X containing x if x X, and otherwise. Definition 1. The binary attribute thinning (non-increasing grain filter) Φ T of set X with criterion T is given by Φ T (X) = x X Φ T (Γ x (X)) (1) A multi-variate attribute thinning Φ {Ti} (X) with scalar attributes {τ i } and their corresponding criteria {T i }, with 1 i N, preserves a component C if i : T i, T i = τ i (C) r i : N Φ {Ti} (X) = Φ T i (X). (2) i=1

6 Binary vector-attribute thinning C is preserved if τ(c) Υ satisfies criterion T r,ɛ τ (C) = d( τ(c), r) ɛ. Dissimilarity measure d : Υ Υ R quantifies the difference between τ(c) and r. A binary vector-attribute thinning Φ r,ɛ τ (X), with D-dimensional vectors from a space Υ R D, removes the connected components of a binary image X whose vector-attributes differ less than ɛ from a reference vector r Υ. Definition 2. The vector-attribute thinning Φ r,ɛ τ of X with respect to a reference vector r and using vector-attribute τ and scalar value ɛ is given by Φ τ r,ɛ(x) = {x X T τ r,ɛ(γ x (X))}. (3) Possible choice for d: Euclidean distance d( u, v) = v u. Any dissimilarity measure can be used (such as Mahalanobis distance). Since the triangle inequality d(a, c) d(a, b) + d(b, c) is not required, d need not be a distance.

7 Gray-scale vector-attribute thinning Extension to gray-scale using threshold decomposition: φ τ r,ɛ(f) = sup{h T τ r,ɛ(γ x (X h (f)))}, (4) where threshold set X h (f) is defined as: X h (f) = {x M f(x) h}. Example: removing letters from image f consisting of nested versions of the letters A, B, and C. f φ τ S A,ɛ (f) φ τ S B,ɛ (f) φ τ S C,ɛ (f) Note: efficient algorithms (such as Max-tree) exists to compute gray-scale attribute thinnings directly without using the expensive threshold decomposition.

8 Moment invariants as vector-attribute Hu s set of seven moment invariants are invariant to rotation, scaling and translation. Krawtchouk moment invariant form a set of discrete and orthogonal moment invariants. Flusser and Suk developed a set of complete and independent moment invariants Problem of Krawtchouk moment invariants when the reference shape is not rotationally symmetric: angle used here is the orientation instead of the direction of the shape

9 Moment invariants as vector-attribute Influence of orientation on dissimilarity using moment invariants of Hu and Krawtchouk: rotated versions of letter A are compared with original A, double-sized A, half-sized A, and letter B (dotted line): Error A Larger A Smaller A B Rotation angle (degrees) Error A Larger A Smaller A B Rotation angle (degrees) Error A Larger A Smaller A B Rotation angle (degrees) Hu Krawtchouk Krawtchouk d( τ(c i ), τ(s 1 )) d( τ(c i ), τ(s 1 )) min 2 n=1 d( τ(c i ), τ(s n )) Note that the letter B should always be more dissimilar to A than any rotated or scaled A. Furthermore, rotation- and scale-invariance is only approximated in the digital case.

10 Hu s Moment invariants Using vector-attribute thinning with Hu s set of 7 moment invariants as vector-attribute to remove from image X the letters A, B, and C respectively. X Φ τ S A,0.010 (X) Φ τ S B,0.013 (X) Φ τ S C,0.010 (X) X Φ τ S A,ɛ (X) X Φ τ S B,ɛ (X) X Φ τ S C,ɛ (X)

11 Pattern spectrum Definition 3. A binary shape granulometry is a set of operators (thinnings) {β r } with r from some totally ordered set Λ, with the following three properties for all r, s Λ and λ > 0. anti-extensiveness: β r (X) X (5) scale-invariance: β r (X λ ) =(β r (X)) λ (6) nesting relationship: β r (β s (X)) =β max(r,s) (X), (7) The (shape) pattern spectrum is defined for a granulometry {β r }: (s β (X))(u) = da(β r(x)) dr, (8) r=u where A(X) denotes the Lebesgue measure in R n, which is the area if n = 2.

12 Pattern spectrum Shape granulometry using shape family F r, which is a set containing the reference shapes of the first r letters of the alphabet. A shape is removed from image X if it resembles at least one of the reference shapes. Amount of detail removed 9 x Number of objects removed Number of letters in F Number of letters in F X Pattern spectrum Shape histogram Y 1 = X Φ τ F 1,ɛ Y 2 = Y 1 Φ τ F 2,ɛ Y 3 = Y 2 Φ τ F 3,ɛ Y 4 = Y 3 Φ τ F 4,ɛ Y 5 = Y 4 Φ τ F 5,ɛ

13 Conclusions New class of attribute filters and granulometries whose attributes are vector instead of scalar values. Attribute thinnings were implemented using Salembier s Max-tree algorithm. Other algorithms for connected (gray-scale) thinnings, such as level line trees, can also be used. Alternative moment invariants such as the complex moment invariants of Flusser and Suk should be investigated. More research is also needed to determine better ways for selecting the parameters like ɛ and the order and the choice of shape classes. Other dissimilarity measures than the Euclidean distance should be investigated (adaptive system such as genetic algorithm, Mahalanobis distance only if multiple reference instances of the target class are available). Because only examples of the target class are used, the filtering problem resembles one-class classification (kernel density estimates). Support-vector domain description could be used in a similar way.

14 Questions

15 Properties of operators Below, the following terms will be used to describe the properties of an operator Ψ for binary images X and Y : idempotence : Ψ(Ψ(X)) = Ψ(X) (9) increasingness : X Y Ψ(X) Ψ(Y ) (10) anti-extensiveness : Ψ(X) X. (11) An idempotent operator is also known as a filter. Operators that are anti-extensive, idempotent, and increasing are called openings. Operators that are anti-extensive, idempotent, but not necessarily increasing are called thinnings. Attribute thinning Pattern spectrum

16 Max-tree A peak component P h of image f is a connected region in which f(x) h for all x P h, and all neighbours of P h have gray level smaller than h. P 0 3 P 0 2 P 1 2 P 0 1 P C 0 3 C 0 2 C 0 1 C 0 0 C 1 2 Peak components Attributes Max-tree Filtering using Max-tree with 4 different rules: P 0 1 P 0 0 P 0 3 P 0 2 P 0 1 P 0 0 P 0 3 P 0 1 P 0 0 P 0 2 P 0 1 P 0 0 Min Max Direct Subtractive Back

17 Hu s moment invariants (Central) moments up to some order (p + q) are computed: Moments: m pq = x p y q f(x, y) dx dy (12) R 2 Central moments: µ pq = (x x) p (y ȳ) q f(x, y) dx dy R 2 (13) Normalized central moments: where x = m 10 m 00 and ȳ = m 01 m 00 (14) η pq = µ pq µ γ 00 where γ = p + q 2 (15) + 1 (16) (17) Back

18 Hu s moment invariants Hu s set of seven moment invariants is defined as: φ 1 =η 20 + η 02 φ 2 =(η 20 η 02 ) 2 + 4η 2 11 (18) (19) φ 3 =(η 30 3η 12 ) 2 + (3η 21 η 03 ) 2 (20) φ 4 =(η 30 + η 12 ) 2 + (η 21 + η 03 ) 2 (21) φ 5 =(η 30 3η 12 )(η 30 + η 12 )[(η 30 + η 12 ) 2 3(η 21 + η 03 ) 2 ] + (3η 21 η 03 )(η 21 + η 03 )[3(η 30 + η 12 ) 2 (η 21 + η 03 ) 2 ] (22) φ 6 =(η 20 η 02 )[(η 30 + η 12 ) 2 (η 21 + η 03 ) 2 ] + 4η 11 (η 30 + η 12 )(η 21 + η 03 ) (23) φ 7 =(3η 21 η 03 )(η 30 + η 12 )[(η 30 + η 12 ) 2 3(η 21 + η 03 ) 2 ] + (3η 12 η 30 )(η 21 + η 03 )[3(η 30 + η 12 ) 2 (η 21 + η 03 ) 2 ] (24) Note that these seven moment invariants are computed using central moments up-to(and including) order 3. Back

19 Krawtchouk moment invariants ν nm = N 1 x=0 N 1 y=0 N 2 2M 00 f(x, y) (25) [(x x) cos θ + (y (y) N sin θ 2 /2 + N M 00 2 [(x x) cos θ + (y (y) N sin θ 2 /2 + N M 00 2 n n where N is the width of the input image and θ = 1 2 tan( 1) 2µ 11 µ 20 µ 02 (26) Back

20 Krawtchouk moment invariants K n (x P, N) = N a k,n,p x k = 2 F 1 ( n, x, N; 1 p ) (27) k=0 where x, n = 0, 1, 2...N, N > 0 and p (0, 1) 2F 1 (a, b, c; z) = k=0 (a) k (b) k (c) k z k k! (28) (a) k = a(a + 1)...(a + k 1) (29) The Krawtchouk moment invariants can now be defined by: Q nm = [ρ(n), ρ(m)] 1 2 n i=0 m a i,n,p1 a j,m,p2 ν ij (30) j=0 where a k,n,p are coefficients determined by equation 27. Back

Rotation-invariant connected size-shape pattern spectra

Rotation-invariant connected size-shape pattern spectra Rotation-invariant connected size-shape pattern spectra Erik R. Urbach Institute for Mathematics and Computing Science University of Groningen The Netherlands April 28, 2004 BCN Retraite Research topic

More information

Differential Attribute Profiles in Remote Sensing

Differential Attribute Profiles in Remote Sensing Differential Attribute Profiles in Remote Sensing Georgios K. Ouzounis DigitalGlobe, Inc. 1601 Dry Creek Drive, Suite 260, Longmont, CO 80503, USA. Image: Dubai July 5, 2011 50cm Contents Overview Connected

More information

Computer Assisted Image Analysis

Computer Assisted Image Analysis Computer Assisted Image Analysis Lecture 0 - Object Descriptors II Amin Allalou amin@cb.uu.se Centre for Image Analysis Uppsala University 2009-04-27 A. Allalou (Uppsala University) Object Descriptors

More information

Chapter I : Basic Notions

Chapter I : Basic Notions Chapter I : Basic Notions - Image Processing - Lattices - Residues - Measurements J. Serra Ecole des Mines de Paris ( 2000 ) Course on Math. Morphology I. 1 Image Processing (I) => Image processing may

More information

(1) Recap of Differential Calculus and Integral Calculus (2) Preview of Calculus in three dimensional space (3) Tools for Calculus 3

(1) Recap of Differential Calculus and Integral Calculus (2) Preview of Calculus in three dimensional space (3) Tools for Calculus 3 Math 127 Introduction and Review (1) Recap of Differential Calculus and Integral Calculus (2) Preview of Calculus in three dimensional space (3) Tools for Calculus 3 MATH 127 Introduction to Calculus III

More information

Linear Algebra Review

Linear Algebra Review January 29, 2013 Table of contents Metrics Metric Given a space X, then d : X X R + 0 and z in X if: d(x, y) = 0 is equivalent to x = y d(x, y) = d(y, x) d(x, y) d(x, z) + d(z, y) is a metric is for all

More information

be the set of complex valued 2π-periodic functions f on R such that

be the set of complex valued 2π-periodic functions f on R such that . Fourier series. Definition.. Given a real number P, we say a complex valued function f on R is P -periodic if f(x + P ) f(x) for all x R. We let be the set of complex valued -periodic functions f on

More information

The Cross Product. MATH 311, Calculus III. J. Robert Buchanan. Fall Department of Mathematics. J. Robert Buchanan The Cross Product

The Cross Product. MATH 311, Calculus III. J. Robert Buchanan. Fall Department of Mathematics. J. Robert Buchanan The Cross Product The Cross Product MATH 311, Calculus III J. Robert Buchanan Department of Mathematics Fall 2011 Introduction Recall: the dot product of two vectors is a scalar. There is another binary operation on vectors

More information

NOTES ON BARNSLEY FERN

NOTES ON BARNSLEY FERN NOTES ON BARNSLEY FERN ERIC MARTIN 1. Affine transformations An affine transformation on the plane is a mapping T that preserves collinearity and ratios of distances: given two points A and B, if C is

More information

Part I : Bases. Part I : Bases

Part I : Bases. Part I : Bases Indian Statistical Institute System Science and Informatics Unit Bengalore, India Bengalore, 19-22 October 2010 ESIEE University of Paris-Est France Part I : Bases - ordering and lattices - erosion and

More information

Course 212: Academic Year Section 1: Metric Spaces

Course 212: Academic Year Section 1: Metric Spaces Course 212: Academic Year 1991-2 Section 1: Metric Spaces D. R. Wilkins Contents 1 Metric Spaces 3 1.1 Distance Functions and Metric Spaces............. 3 1.2 Convergence and Continuity in Metric Spaces.........

More information

Functional Analysis Review

Functional Analysis Review Outline 9.520: Statistical Learning Theory and Applications February 8, 2010 Outline 1 2 3 4 Vector Space Outline A vector space is a set V with binary operations +: V V V and : R V V such that for all

More information

Chapter VIII : Thinnings

Chapter VIII : Thinnings Chapter VIII : Thinnings Hit-or-Miss Thinning thickening Homotopy Homotopy and Connectivity Homotopic Thinnings and Thickenings J. Serra Ecole des Mines de Paris ( 2000 ) Course on Math. Morphology VIII.

More information

Reflections and Rotations in R 3

Reflections and Rotations in R 3 Reflections and Rotations in R 3 P. J. Ryan May 29, 21 Rotations as Compositions of Reflections Recall that the reflection in the hyperplane H through the origin in R n is given by f(x) = x 2 ξ, x ξ (1)

More information

Representing regions in 2 ways:

Representing regions in 2 ways: Representing regions in 2 ways: Based on their external characteristics (its boundary): Shape characteristics Based on their internal characteristics (its region): Both Regional properties: color, texture,

More information

15B. Isometries and Functionals 1

15B. Isometries and Functionals 1 5B. Isometries and Functionals Isometries If the important functions between vector spaces V and W are those that preserve linearity (i.e., linear transformations), then the important functions between

More information

(x, y) = d(x, y) = x y.

(x, y) = d(x, y) = x y. 1 Euclidean geometry 1.1 Euclidean space Our story begins with a geometry which will be familiar to all readers, namely the geometry of Euclidean space. In this first chapter we study the Euclidean distance

More information

Omm Al-Qura University Dr. Abdulsalam Ai LECTURE OUTLINE CHAPTER 3. Vectors in Physics

Omm Al-Qura University Dr. Abdulsalam Ai LECTURE OUTLINE CHAPTER 3. Vectors in Physics LECTURE OUTLINE CHAPTER 3 Vectors in Physics 3-1 Scalars Versus Vectors Scalar a numerical value (number with units). May be positive or negative. Examples: temperature, speed, height, and mass. Vector

More information

Chapter 6. Differentially Flat Systems

Chapter 6. Differentially Flat Systems Contents CAS, Mines-ParisTech 2008 Contents Contents 1, Linear Case Introductory Example: Linear Motor with Appended Mass General Solution (Linear Case) Contents Contents 1, Linear Case Introductory Example:

More information

Vectors in Function Spaces

Vectors in Function Spaces Jim Lambers MAT 66 Spring Semester 15-16 Lecture 18 Notes These notes correspond to Section 6.3 in the text. Vectors in Function Spaces We begin with some necessary terminology. A vector space V, also

More information

Linear Models Review

Linear Models Review Linear Models Review Vectors in IR n will be written as ordered n-tuples which are understood to be column vectors, or n 1 matrices. A vector variable will be indicted with bold face, and the prime sign

More information

2. A die is rolled 3 times, the probability of getting a number larger than the previous number each time is

2. A die is rolled 3 times, the probability of getting a number larger than the previous number each time is . If P(A) = x, P = 2x, P(A B) = 2, P ( A B) = 2 3, then the value of x is (A) 5 8 5 36 6 36 36 2. A die is rolled 3 times, the probability of getting a number larger than the previous number each time

More information

Properties of Matrices and Operations on Matrices

Properties of Matrices and Operations on Matrices Properties of Matrices and Operations on Matrices A common data structure for statistical analysis is a rectangular array or matris. Rows represent individual observational units, or just observations,

More information

a Write down the coordinates of the point on the curve where t = 2. b Find the value of t at the point on the curve with coordinates ( 5 4, 8).

a Write down the coordinates of the point on the curve where t = 2. b Find the value of t at the point on the curve with coordinates ( 5 4, 8). Worksheet A 1 A curve is given by the parametric equations x = t + 1, y = 4 t. a Write down the coordinates of the point on the curve where t =. b Find the value of t at the point on the curve with coordinates

More information

Newtonian Mechanics. Chapter Classical space-time

Newtonian Mechanics. Chapter Classical space-time Chapter 1 Newtonian Mechanics In these notes classical mechanics will be viewed as a mathematical model for the description of physical systems consisting of a certain (generally finite) number of particles

More information

Continuous Ordinal Clustering: A Mystery Story 1

Continuous Ordinal Clustering: A Mystery Story 1 Continuous Ordinal Clustering: A Mystery Story 1 Melvin F. Janowitz Abstract Cluster analysis may be considered as an aid to decision theory because of its ability to group the various alternatives. There

More information

Chapter VI. Set Connection and Numerical Functions

Chapter VI. Set Connection and Numerical Functions Chapter VI Set Connection and Numerical Functions Concepts : -> Extension to functions -> Connected operators -> Numerical Geodesy -> Leveling and self-duality Applications : -> Extrema analysis -> Contour

More information

2.1 Scalars and Vectors

2.1 Scalars and Vectors 2.1 Scalars and Vectors Scalar A quantity characterized by a positive or negative number Indicated by letters in italic such as A e.g. Mass, volume and length 2.1 Scalars and Vectors Vector A quantity

More information

Math 4317 : Real Analysis I Mid-Term Exam 1 25 September 2012

Math 4317 : Real Analysis I Mid-Term Exam 1 25 September 2012 Instructions: Answer all of the problems. Math 4317 : Real Analysis I Mid-Term Exam 1 25 September 2012 Definitions (2 points each) 1. State the definition of a metric space. A metric space (X, d) is set

More information

EGR2013 Tutorial 8. Linear Algebra. Powers of a Matrix and Matrix Polynomial

EGR2013 Tutorial 8. Linear Algebra. Powers of a Matrix and Matrix Polynomial EGR1 Tutorial 8 Linear Algebra Outline Powers of a Matrix and Matrix Polynomial Vector Algebra Vector Spaces Powers of a Matrix and Matrix Polynomial If A is a square matrix, then we define the nonnegative

More information

Mathematics Review Exercises. (answers at end)

Mathematics Review Exercises. (answers at end) Brock University Physics 1P21/1P91 Mathematics Review Exercises (answers at end) Work each exercise without using a calculator. 1. Express each number in scientific notation. (a) 437.1 (b) 563, 000 (c)

More information

Vectors. The standard geometric definition of vector is as something which has direction and magnitude but not position.

Vectors. The standard geometric definition of vector is as something which has direction and magnitude but not position. Vectors The standard geometric definition of vector is as something which has direction and magnitude but not position. Since vectors have no position we may place them wherever is convenient. Vectors

More information

MORPHOLOGICAL IMAGE ANALYSIS III

MORPHOLOGICAL IMAGE ANALYSIS III MORPHOLOGICAL IMAGE ANALYSIS III Secondary inary Morphological Operators A set operator Ψ( ) which is: Increasing; i.e., Ψ( ) Ψ( ) Anti-extensive; i.e., Ψ( ) 1 2 1 2 Idempotent; i.e., ΨΨ ( ( )) = Ψ( )

More information

Densities for the Navier Stokes equations with noise

Densities for the Navier Stokes equations with noise Densities for the Navier Stokes equations with noise Marco Romito Università di Pisa Universitat de Barcelona March 25, 2015 Summary 1 Introduction & motivations 2 Malliavin calculus 3 Besov bounds 4 Other

More information

APPENDIX A. Background Mathematics. A.1 Linear Algebra. Vector algebra. Let x denote the n-dimensional column vector with components x 1 x 2.

APPENDIX A. Background Mathematics. A.1 Linear Algebra. Vector algebra. Let x denote the n-dimensional column vector with components x 1 x 2. APPENDIX A Background Mathematics A. Linear Algebra A.. Vector algebra Let x denote the n-dimensional column vector with components 0 x x 2 B C @. A x n Definition 6 (scalar product). The scalar product

More information

Review of Coordinate Systems

Review of Coordinate Systems Vector in 2 R and 3 R Review of Coordinate Systems Used to describe the position of a point in space Common coordinate systems are: Cartesian Polar Cartesian Coordinate System Also called rectangular coordinate

More information

THE use of moment invariants in object recognition is

THE use of moment invariants in object recognition is IEEE TRANSACTIONS ON IMAGE PROCESSING 1 Comparison of Image Patches Using Local Moment Invariants Atilla Sit and Daisuke Kihara Abstract We propose a new set of moment invariants based on Krawtchouk polynomials

More information

Solutions: Problem Set 4 Math 201B, Winter 2007

Solutions: Problem Set 4 Math 201B, Winter 2007 Solutions: Problem Set 4 Math 2B, Winter 27 Problem. (a Define f : by { x /2 if < x

More information

Math General Topology Fall 2012 Homework 1 Solutions

Math General Topology Fall 2012 Homework 1 Solutions Math 535 - General Topology Fall 2012 Homework 1 Solutions Definition. Let V be a (real or complex) vector space. A norm on V is a function : V R satisfying: 1. Positivity: x 0 for all x V and moreover

More information

2.2. OPERATOR ALGEBRA 19. If S is a subset of E, then the set

2.2. OPERATOR ALGEBRA 19. If S is a subset of E, then the set 2.2. OPERATOR ALGEBRA 19 2.2 Operator Algebra 2.2.1 Algebra of Operators on a Vector Space A linear operator on a vector space E is a mapping L : E E satisfying the condition u, v E, a R, L(u + v) = L(u)

More information

GEOMETRY AND VECTORS

GEOMETRY AND VECTORS GEOMETRY AND VECTORS Distinguishing Between Points in Space One Approach Names: ( Fred, Steve, Alice...) Problem: distance & direction must be defined point-by-point More elegant take advantage of geometry

More information

Mathematical Tripos Part IA Lent Term Example Sheet 1. Calculate its tangent vector dr/du at each point and hence find its total length.

Mathematical Tripos Part IA Lent Term Example Sheet 1. Calculate its tangent vector dr/du at each point and hence find its total length. Mathematical Tripos Part IA Lent Term 205 ector Calculus Prof B C Allanach Example Sheet Sketch the curve in the plane given parametrically by r(u) = ( x(u), y(u) ) = ( a cos 3 u, a sin 3 u ) with 0 u

More information

A matrix over a field F is a rectangular array of elements from F. The symbol

A matrix over a field F is a rectangular array of elements from F. The symbol Chapter MATRICES Matrix arithmetic A matrix over a field F is a rectangular array of elements from F The symbol M m n (F ) denotes the collection of all m n matrices over F Matrices will usually be denoted

More information

MAT 771 FUNCTIONAL ANALYSIS HOMEWORK 3. (1) Let V be the vector space of all bounded or unbounded sequences of complex numbers.

MAT 771 FUNCTIONAL ANALYSIS HOMEWORK 3. (1) Let V be the vector space of all bounded or unbounded sequences of complex numbers. MAT 771 FUNCTIONAL ANALYSIS HOMEWORK 3 (1) Let V be the vector space of all bounded or unbounded sequences of complex numbers. (a) Define d : V V + {0} by d(x, y) = 1 ξ j η j 2 j 1 + ξ j η j. Show that

More information

Block Coordinate Descent for Regularized Multi-convex Optimization

Block Coordinate Descent for Regularized Multi-convex Optimization Block Coordinate Descent for Regularized Multi-convex Optimization Yangyang Xu and Wotao Yin CAAM Department, Rice University February 15, 2013 Multi-convex optimization Model definition Applications Outline

More information

Math 426H (Differential Geometry) Final Exam April 24, 2006.

Math 426H (Differential Geometry) Final Exam April 24, 2006. Math 426H Differential Geometry Final Exam April 24, 6. 8 8 8 6 1. Let M be a surface and let : [0, 1] M be a smooth loop. Let φ be a 1-form on M. a Suppose φ is exact i.e. φ = df for some f : M R. Show

More information

A new class of shift-invariant operators

A new class of shift-invariant operators 1 A new class of shift-invariant operators Janne Heiilä Machine Vision Group Department of Electrical and Information Engineering P.O. Box 4500, 90014 University of Oulu, Finland Tel.: +358 8 553 2786,

More information

Inner Product Spaces

Inner Product Spaces Inner Product Spaces Linear Algebra Josh Engwer TTU 28 October 2015 Josh Engwer (TTU) Inner Product Spaces 28 October 2015 1 / 15 Inner Product Space (Definition) An inner product is the notion of a dot

More information

MA Spring 2013 Lecture Topics

MA Spring 2013 Lecture Topics LECTURE 1 Chapter 12.1 Coordinate Systems Chapter 12.2 Vectors MA 16200 Spring 2013 Lecture Topics Let a,b,c,d be constants. 1. Describe a right hand rectangular coordinate system. Plot point (a,b,c) inn

More information

Analysis Qualifying Exam

Analysis Qualifying Exam Analysis Qualifying Exam Spring 2017 Problem 1: Let f be differentiable on R. Suppose that there exists M > 0 such that f(k) M for each integer k, and f (x) M for all x R. Show that f is bounded, i.e.,

More information

FUNCTIONAL ANALYSIS-NORMED SPACE

FUNCTIONAL ANALYSIS-NORMED SPACE MAT641- MSC Mathematics, MNIT Jaipur FUNCTIONAL ANALYSIS-NORMED SPACE DR. RITU AGARWAL MALAVIYA NATIONAL INSTITUTE OF TECHNOLOGY JAIPUR 1. Normed space Norm generalizes the concept of length in an arbitrary

More information

2. Linear algebra. matrices and vectors. linear equations. range and nullspace of matrices. function of vectors, gradient and Hessian

2. Linear algebra. matrices and vectors. linear equations. range and nullspace of matrices. function of vectors, gradient and Hessian FE661 - Statistical Methods for Financial Engineering 2. Linear algebra Jitkomut Songsiri matrices and vectors linear equations range and nullspace of matrices function of vectors, gradient and Hessian

More information

Assignment 2 : Probabilistic Methods

Assignment 2 : Probabilistic Methods Assignment 2 : Probabilistic Methods jverstra@math Question 1. Let d N, c R +, and let G n be an n-vertex d-regular graph. Suppose that vertices of G n are selected independently with probability p, where

More information

Chapter Objectives. Copyright 2011 Pearson Education South Asia Pte Ltd

Chapter Objectives. Copyright 2011 Pearson Education South Asia Pte Ltd Chapter Objectives To show how to add forces and resolve them into components using the Parallelogram Law. To express force and position in Cartesian vector form and explain how to determine the vector

More information

16.2 Iterated Integrals

16.2 Iterated Integrals 6.2 Iterated Integrals So far: We have defined what we mean by a double integral. We have estimated the value of a double integral from contour diagrams and from tables of values. We have interpreted the

More information

17.1 Hyperplanes and Linear Forms

17.1 Hyperplanes and Linear Forms This is page 530 Printer: Opaque this 17 Appendix 17.1 Hyperplanes and Linear Forms Given a vector space E over a field K, a linear map f: E K is called a linear form. The set of all linear forms f: E

More information

49. Green s Theorem. The following table will help you plan your calculation accordingly. C is a simple closed loop 0 Use Green s Theorem

49. Green s Theorem. The following table will help you plan your calculation accordingly. C is a simple closed loop 0 Use Green s Theorem 49. Green s Theorem Let F(x, y) = M(x, y), N(x, y) be a vector field in, and suppose is a path that starts and ends at the same point such that it does not cross itself. Such a path is called a simple

More information

Camera Models and Affine Multiple Views Geometry

Camera Models and Affine Multiple Views Geometry Camera Models and Affine Multiple Views Geometry Subhashis Banerjee Dept. Computer Science and Engineering IIT Delhi email: suban@cse.iitd.ac.in May 29, 2001 1 1 Camera Models A Camera transforms a 3D

More information

Chapter SSM: Linear Algebra Section Fails to be invertible; since det = 6 6 = Invertible; since det = = 2.

Chapter SSM: Linear Algebra Section Fails to be invertible; since det = 6 6 = Invertible; since det = = 2. SSM: Linear Algebra Section 61 61 Chapter 6 1 2 1 Fails to be invertible; since det = 6 6 = 0 3 6 3 5 3 Invertible; since det = 33 35 = 2 7 11 5 Invertible; since det 2 5 7 0 11 7 = 2 11 5 + 0 + 0 0 0

More information

Euclidean Spaces. Euclidean Spaces. Chapter 10 -S&B

Euclidean Spaces. Euclidean Spaces. Chapter 10 -S&B Chapter 10 -S&B The Real Line: every real number is represented by exactly one point on the line. The plane (i.e., consumption bundles): Pairs of numbers have a geometric representation Cartesian plane

More information

Lecture Notes in Advanced Calculus 1 (80315) Raz Kupferman Institute of Mathematics The Hebrew University

Lecture Notes in Advanced Calculus 1 (80315) Raz Kupferman Institute of Mathematics The Hebrew University Lecture Notes in Advanced Calculus 1 (80315) Raz Kupferman Institute of Mathematics The Hebrew University February 7, 2007 2 Contents 1 Metric Spaces 1 1.1 Basic definitions...........................

More information

Linear Algebra. Alvin Lin. August December 2017

Linear Algebra. Alvin Lin. August December 2017 Linear Algebra Alvin Lin August 207 - December 207 Linear Algebra The study of linear algebra is about two basic things. We study vector spaces and structure preserving maps between vector spaces. A vector

More information

Chapter III: Opening, Closing

Chapter III: Opening, Closing Chapter III: Opening, Closing Opening and Closing by adjunction Algebraic Opening and Closing Top-Hat Transformation Granulometry J. Serra Ecole des Mines de Paris ( 2000 ) Course on Math. Morphology III.

More information

Instructions: No books. No notes. Non-graphing calculators only. You are encouraged, although not required, to show your work.

Instructions: No books. No notes. Non-graphing calculators only. You are encouraged, although not required, to show your work. Exam 3 Math 850-007 Fall 04 Odenthal Name: Instructions: No books. No notes. Non-graphing calculators only. You are encouraged, although not required, to show your work.. Evaluate the iterated integral

More information

Analytic Geometry. Orthogonal projection. Chapter 4 Matrix decomposition

Analytic Geometry. Orthogonal projection. Chapter 4 Matrix decomposition 1541 3 Analytic Geometry 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 In Chapter 2, we studied vectors, vector spaces and linear mappings at a general but abstract level. In this

More information

Region Description for Recognition

Region Description for Recognition Region Description for Recognition For object recognition, descriptions of regions in an image have to be compared with descriptions of regions of meaningful objects (models). The general problem of object

More information

PHYS 301 FIRST HOUR EXAM SPRING 2012

PHYS 301 FIRST HOUR EXAM SPRING 2012 PHY 301 FIRT HOUR EXAM PRING 2012 This is a closed book, closed note exam. Turn off and store out of sight all electronic devices for the duration of the exam. Do all your writing in your blue book(s)

More information

2. Review of Linear Algebra

2. Review of Linear Algebra 2. Review of Linear Algebra ECE 83, Spring 217 In this course we will represent signals as vectors and operators (e.g., filters, transforms, etc) as matrices. This lecture reviews basic concepts from linear

More information

The moment-lp and moment-sos approaches

The moment-lp and moment-sos approaches The moment-lp and moment-sos approaches LAAS-CNRS and Institute of Mathematics, Toulouse, France CIRM, November 2013 Semidefinite Programming Why polynomial optimization? LP- and SDP- CERTIFICATES of POSITIVITY

More information

4sec 2xtan 2x 1ii C3 Differentiation trig

4sec 2xtan 2x 1ii C3 Differentiation trig A Assignment beta Cover Sheet Name: Question Done Backpack Topic Comment Drill Consolidation i C3 Differentiation trig 4sec xtan x ii C3 Differentiation trig 6cot 3xcosec 3x iii C3 Differentiation trig

More information

Mathematics skills framework

Mathematics skills framework Mathematics skills framework The framework for MYP mathematics outlines four branches of mathematical study. Schools can use the framework for mathematics as a tool for curriculum mapping when designing

More information

Value Iteration and Action ɛ-approximation of Optimal Policies in Discounted Markov Decision Processes

Value Iteration and Action ɛ-approximation of Optimal Policies in Discounted Markov Decision Processes Value Iteration and Action ɛ-approximation of Optimal Policies in Discounted Markov Decision Processes RAÚL MONTES-DE-OCA Departamento de Matemáticas Universidad Autónoma Metropolitana-Iztapalapa San Rafael

More information

The 2D orientation is unique through Principal Moments Analysis

The 2D orientation is unique through Principal Moments Analysis The 2D orientation is unique through Principal Moments Analysis João F. P. Crespo Pedro M. Q. Aguiar Institute for Systems and Robotics / IST Lisboa, Portugal September, 2010 Outline 1 Motivation 2 Moments

More information

Blow-up on manifolds with symmetry for the nonlinear Schröding

Blow-up on manifolds with symmetry for the nonlinear Schröding Blow-up on manifolds with symmetry for the nonlinear Schrödinger equation March, 27 2013 Université de Nice Euclidean L 2 -critical theory Consider the one dimensional equation i t u + u = u 4 u, t > 0,

More information

Affine Normalization of Symmetric Objects

Affine Normalization of Symmetric Objects Affine Normalization of Symmetric Objects Tomáš Suk and Jan Flusser Institute of Information Theory and Automation, Academy of Sciences of the Czech Republic, Pod vodárenskou věží 4, 182 08 Prague 8, Czech

More information

u xx + u yy = 0. (5.1)

u xx + u yy = 0. (5.1) Chapter 5 Laplace Equation The following equation is called Laplace equation in two independent variables x, y: The non-homogeneous problem u xx + u yy =. (5.1) u xx + u yy = F, (5.) where F is a function

More information

Math 113/114 Lecture 22

Math 113/114 Lecture 22 Math 113/114 Lecture 22 Xi Chen 1 1 University of Alberta October 31, 2014 Outline 1 2 (Application of Implicit Differentiation) Given a word problem about related rates, we need to do: interpret the problem

More information

Vectors in the Plane

Vectors in the Plane Vectors in the Plane MATH 311, Calculus III J. Robert Buchanan Department of Mathematics Fall 2011 Vectors vs. Scalars scalar quantity having only a magnitude (e.g. temperature, volume, length, area) and

More information

Introduction and Vectors Lecture 1

Introduction and Vectors Lecture 1 1 Introduction Introduction and Vectors Lecture 1 This is a course on classical Electromagnetism. It is the foundation for more advanced courses in modern physics. All physics of the modern era, from quantum

More information

UNIVERSITY OF BRISTOL. Mock exam paper for examination for the Degrees of B.Sc. and M.Sci. (Level 3)

UNIVERSITY OF BRISTOL. Mock exam paper for examination for the Degrees of B.Sc. and M.Sci. (Level 3) UNIVERSITY OF BRISTOL Mock exam paper for examination for the Degrees of B.Sc. and M.Sci. (Level 3) DYNAMICAL SYSTEMS and ERGODIC THEORY MATH 36206 (Paper Code MATH-36206) 2 hours and 30 minutes This paper

More information

Chapter 2 Statics of Particles. Resultant of Two Forces 8/28/2014. The effects of forces on particles:

Chapter 2 Statics of Particles. Resultant of Two Forces 8/28/2014. The effects of forces on particles: Chapter 2 Statics of Particles The effects of forces on particles: - replacing multiple forces acting on a particle with a single equivalent or resultant force, - relations between forces acting on a particle

More information

Section 3.9. Matrix Norm

Section 3.9. Matrix Norm 3.9. Matrix Norm 1 Section 3.9. Matrix Norm Note. We define several matrix norms, some similar to vector norms and some reflecting how multiplication by a matrix affects the norm of a vector. We use matrix

More information

Vectors for Physics. AP Physics C

Vectors for Physics. AP Physics C Vectors for Physics AP Physics C A Vector is a quantity that has a magnitude (size) AND a direction. can be in one-dimension, two-dimensions, or even three-dimensions can be represented using a magnitude

More information

Linear Algebra (Review) Volker Tresp 2017

Linear Algebra (Review) Volker Tresp 2017 Linear Algebra (Review) Volker Tresp 2017 1 Vectors k is a scalar (a number) c is a column vector. Thus in two dimensions, c = ( c1 c 2 ) (Advanced: More precisely, a vector is defined in a vector space.

More information

Volume: The Disk Method. Using the integral to find volume.

Volume: The Disk Method. Using the integral to find volume. Volume: The Disk Method Using the integral to find volume. If a region in a plane is revolved about a line, the resulting solid is a solid of revolution and the line is called the axis of revolution. y

More information

Data Mining: Data. Lecture Notes for Chapter 2. Introduction to Data Mining

Data Mining: Data. Lecture Notes for Chapter 2. Introduction to Data Mining Data Mining: Data Lecture Notes for Chapter 2 Introduction to Data Mining by Tan, Steinbach, Kumar Similarity and Dissimilarity Similarity Numerical measure of how alike two data objects are. Is higher

More information

Physics 411 Lecture 7. Tensors. Lecture 7. Physics 411 Classical Mechanics II

Physics 411 Lecture 7. Tensors. Lecture 7. Physics 411 Classical Mechanics II Physics 411 Lecture 7 Tensors Lecture 7 Physics 411 Classical Mechanics II September 12th 2007 In Electrodynamics, the implicit law governing the motion of particles is F α = m ẍ α. This is also true,

More information

= , AD = 1, BE = 1, and F G = 1. By Pythagoras, BD = AB 2 + AD 2 = 2. By similar triangles on DEF and ABD, we have that AF.

= , AD = 1, BE = 1, and F G = 1. By Pythagoras, BD = AB 2 + AD 2 = 2. By similar triangles on DEF and ABD, we have that AF. RMT 0 Team Test Solutions February 0, 0 How many squares are there in the xy-plane such that both coordinates of each vertex are integers between 0 and 00 inclusive, and the sides are parallel to the axes?

More information

c i x (i) =0 i=1 Otherwise, the vectors are linearly independent. 1

c i x (i) =0 i=1 Otherwise, the vectors are linearly independent. 1 Hilbert Spaces Physics 195 Supplementary Notes 009 F. Porter These notes collect and remind you of several definitions, connected with the notion of a Hilbert space. Def: A (nonempty) set V is called a

More information

Contravariant and Covariant as Transforms

Contravariant and Covariant as Transforms Contravariant and Covariant as Transforms There is a lot more behind the concepts of contravariant and covariant tensors (of any rank) than the fact that their basis vectors are mutually orthogonal to

More information

PRELIMINARIES FOR GENERAL TOPOLOGY. Contents

PRELIMINARIES FOR GENERAL TOPOLOGY. Contents PRELIMINARIES FOR GENERAL TOPOLOGY DAVID G.L. WANG Contents 1. Sets 2 2. Operations on sets 3 3. Maps 5 4. Countability of sets 7 5. Others a mathematician knows 8 6. Remarks 9 Date: April 26, 2018. 2

More information

for all subintervals I J. If the same is true for the dyadic subintervals I D J only, we will write ϕ BMO d (J). In fact, the following is true

for all subintervals I J. If the same is true for the dyadic subintervals I D J only, we will write ϕ BMO d (J). In fact, the following is true 3 ohn Nirenberg inequality, Part I A function ϕ L () belongs to the space BMO() if sup ϕ(s) ϕ I I I < for all subintervals I If the same is true for the dyadic subintervals I D only, we will write ϕ BMO

More information

Normed & Inner Product Vector Spaces

Normed & Inner Product Vector Spaces Normed & Inner Product Vector Spaces ECE 174 Introduction to Linear & Nonlinear Optimization Ken Kreutz-Delgado ECE Department, UC San Diego Ken Kreutz-Delgado (UC San Diego) ECE 174 Fall 2016 1 / 27 Normed

More information

The Cross Product. In this section, we will learn about: Cross products of vectors and their applications.

The Cross Product. In this section, we will learn about: Cross products of vectors and their applications. The Cross Product In this section, we will learn about: Cross products of vectors and their applications. THE CROSS PRODUCT The cross product a x b of two vectors a and b, unlike the dot product, is a

More information

FFTs in Graphics and Vision. Groups and Representations

FFTs in Graphics and Vision. Groups and Representations FFTs in Graphics and Vision Groups and Representations Outline Groups Representations Schur s Lemma Correlation Groups A group is a set of elements G with a binary operation (often denoted ) such that

More information

x + ye z2 + ze y2, y + xe z2 + ze x2, z and where T is the

x + ye z2 + ze y2, y + xe z2 + ze x2, z and where T is the 1.(8pts) Find F ds where F = x + ye z + ze y, y + xe z + ze x, z and where T is the T surface in the pictures. (The two pictures are two views of the same surface.) The boundary of T is the unit circle

More information

University of Florida CISE department Gator Engineering. Clustering Part 1

University of Florida CISE department Gator Engineering. Clustering Part 1 Clustering Part 1 Dr. Sanjay Ranka Professor Computer and Information Science and Engineering University of Florida, Gainesville What is Cluster Analysis? Finding groups of objects such that the objects

More information

Information About Ellipses

Information About Ellipses Information About Ellipses David Eberly, Geometric Tools, Redmond WA 9805 https://www.geometrictools.com/ This work is licensed under the Creative Commons Attribution 4.0 International License. To view

More information

Invariances in spectral estimates. Paris-Est Marne-la-Vallée, January 2011

Invariances in spectral estimates. Paris-Est Marne-la-Vallée, January 2011 Invariances in spectral estimates Franck Barthe Dario Cordero-Erausquin Paris-Est Marne-la-Vallée, January 2011 Notation Notation Given a probability measure ν on some Euclidean space, the Poincaré constant

More information

ELEMENTARY LINEAR ALGEBRA

ELEMENTARY LINEAR ALGEBRA ELEMENTARY LINEAR ALGEBRA K R MATTHEWS DEPARTMENT OF MATHEMATICS UNIVERSITY OF QUEENSLAND First Printing, 99 Chapter LINEAR EQUATIONS Introduction to linear equations A linear equation in n unknowns x,

More information