Texture Segmentation

Size: px
Start display at page:

Download "Texture Segmentation"

Transcription

1 Texture Segmentation Introduction to Signal and Image Processing Prof. Dr. Philippe Cattin MIAC, University of Basel 1 of :20

2 Contents Contents Abstract 2 1 Introduction What is Texture? 4 What do we need it for? 5 Characteristics of Textures 6 Principle of Texture Segmentation Methods 7 Problems with Texture Segmentation is Scale 8 2 Feature Extraction Feature Extraction First-Order Histogram Based Features First-Order Histogram Based Features 12 Histogram Features 13 Properties of these Histogram Features 14 First-Order Texture Feature Example Cooccurrence Matrices Cooccurrence Matrices 17 The Cooccurrence Matrix 18 Cooccurrence Matrix Examples 19 Cooccurrence Matrix Examples (2) 20 Extensions to the Cooccurrence Matrix 21 Cooccurrence Matrix Features 22 Haralick Feature Examples Fourier Features Fourier Features 25 Fourier Features: Entropy Example 26 Fourier Features (2) 27 Disadvantage of the Fourier Features Filter Banks Filter Banks Laws Filters Laws Filters 32 2 of :20

3 2.4.2 Gabor Filters Gabor Filters Gabor Filter Power Spectrum Gabor Filter in the Frequency Domain Gabor Filter Example Eigenfilters Eigenfilters 3 Texture Discrimination Generalisation of Eigenfilters Eigenfilter Example Eigenfilter Example - Periodicity of the Texture Eigenfilter Example - Extract the Sample Vectors Eigenfilter Example - Calculation of the Eigenfilters Eigenfilter Example - Eigenfilter the Flawed Textile Eigenfilter Example - Smooth the Filter Example Eigenfilter Example - Mahalanobis Distance Texture Discrimination 3.1 Region Growing Region Growing Methods Region Growing by Pixel Aggregation 3.2 Other Methods Other Methods 3.3 Reduction of Dimensionality Reduction of Dimensionality 4 Texture Classification Texture Classification 5 References References of :20

4 Abstract (2) 4 of :20

5 Introduction What is Texture? (4) The term texture is difficult to define, but it represents aspects of the surface pattern, such as coarseness, directionality, brightness, colour, regularity, and so forth. Texture is a phenomenon that is widespread, easy to recognise and hard to define. Forsyth/Ponce, of :20

6 Introduction What do we need it for? (5) Texture features can be used to segment scenes into parts. In remote sensing it is quite usual to segment scenes into different types of land, such as forrest, vineyards, urban, streets, lakes,... Fig 8.1: Segmentation example (forrest, cultural land, other), ETH Zurich 6 of :20

7 Introduction Characteristics of Textures (6) Already form the small set of textures some differences can be observed (a)+(b) are regular and contain some basic patter (texel) that is repeated in some way (placement rule) (a)'s texel is larger than (b)'s. (a) is thus said to be coarser It is even more subtle, as texture (a) has subpatterns (micro-texture) hierarchical organisation Texture (a) has a dominant diagonal organisation, (b) has a vertical orientation, (c)+(d) have no apparent orientation (a) (b) (c) (d) Fig 8.2: Sample textures 7 of :20

8 Introduction Principle of Texture Segmentation Methods (7) Texture analysis methods typically probe for the aforementioned characteristics like regularity, coarseness, and directionality. Texture analysis can be split in three fundamental steps as shown in Fig Feature extraction: Extract the texture measures, that numerically describe the texture properties 2. Texture discrimination: partition a textured image into regions, each corresponding to a perceptually homogeneous texture 3. Texture classification: determine to which of a finite number of physically defined classes a homogeneous textured region belongs Fig 8.3: Fundamentals steps in texture analysis 8 of :20

9 Introduction Problems with Texture Segmentation is Scale (8) (a) Segment forest, water,... (b) Segment fields, forest, streets,... (c) Segment small groups of trees, grass,... (d) Segment the leafs (e) Segment structure of the leaf (f) Segment the cells 9 of :20

10 Feature Extraction Feature Extraction (10) The approaches to extract the texture measures are typically categorised into 1. statistical, 2. structural, 3. transform, and 4. model-based approaches. Fig 8.4: Typology of the texture measures commonly used in texture analysis 10 of :20

11 First-Order Histogram Based Features First-Order Histogram Based Features Given an image of two space variables and. The function can take the discrete values, where is the number of intensity levels in the image. The intensity level histogram is a function showing the number of pixels in the whole image with this intensity level: (a) (12) (8.1) where is the Kronecker delta function (8.2) (b) Fig 8.5: (a) Texture, (b) corresponding histogram The probability density function (PDF) can be approximated by deviding by the total number of pixels, thus 11 of :20

12 First-Order Histogram Based Features Histogram Features (13) The shape of the histogram can subsequently be used to characterise the image. For example, a narrowly distributed histogram indicates a low-contrast image, and a bimodal histogram often suggests a low contrast object against a differing background. To quantitatively describe the image, different parameters (image features) can be extracted from the histogram. Mean: Skewness: Energy: Variance: Kurtosis: Entropy: 12 of :20

13 First-Order Histogram Based Features Properties of these Histogram Features (14) The mean takes the average level of intensity in the image The variance describes the variation of the intensity around the mean is a measure of contrast The energy is a measure of peakness The skewness is zero if the histogram is symmetrical about the mean indicator of symmetry The kurtosis is a measure of flatness of the histogram for Gaussian shaped histograms The entropy is a measure of histogram uniformity The mean and variance do not actually carry direct information about the texture. They rather depend on the image acquisition process. One often gets better texture segmentation performance if the images are normalised to a mean of and a standard deviation of. 13 of :20

14 First-Order Histogram Based Features First-Order Texture Feature Example (15) Mean Variance (a) Skewness (b) Kurtosis Energy (c) Entropy (d) Fig 8.6: (a) Texture 1, (b) Texture 2, (c) Histogram of texture 1, (d) Histogram of texture 2 14 of :20

15 Cooccurrence Matrices Cooccurrence Matrices (17) The histogram as used in the previous section captures intensity information in a highly compact, but also highly incomplete way. As Fig. 8.7 shows, two textures with a completely different visual appearance can have the same gray-scale histogram. All information about configurational aspects of the intensity distribution is lost. (a) Texture 1 (b) Histogram of texture 1 The Cooccurrence matrix, in contrast, is a related data structure that bases on second-order statistics and preserves some aspects of the spatial configurations. (c) Texture 2 (d) Histogram of texture 2 Fig 8.7: Completely different textures can have the same gray-scale histogram 15 of :20

16 Cooccurrence Matrices The Cooccurrence Matrix (18) The Cooccurrence matrix is in the literature also known as the GLCM (Gray-Level Co-occurrence Matrices), or the Spatial dependence matrix. Mathematically, the cooccurrence matrix is defined over an image, parametrised by an offset as (8.3) The cooccurrence matrix is extracted for a specific vector. Fig 8.8: Construction of the cooccurrence matrix 16 of :20

17 Cooccurrence Matrices Cooccurrence Matrix Examples (19) Fig 8.9: Cooccurrence matrix sample of a texture with a vector ( does correspond to a period of the texture. ) that The high values in the cooccurrence matrix are concentrated in a central location. 17 of :20

18 Cooccurrence Matrices Cooccurrence Matrix Examples (2) (20) Fig 8.10: Cooccurrence matrix sample of a texture with a vector ( ) that does not correspond to a period of the texture. In contrast to the previous example, the values in the cooccurrence matrix are more smeared out. 18 of :20

19 Cooccurrence Matrices Extensions to the Cooccurrence Matrix (21) Several intuitive extension of the Cooccurrence matrix are commonly used in pactice: 1. To analyse a texture, one can build several such matrices, each for a different choice of vector if the length of the vector is varied, a certain degree of scale invariance could be achieved. 2. One can also use several vectors at once and put the results into a single matrix vectors of the same length but with a different orientation combined results in a rotation invariant cooccurrence matrix. Fig 8.11: Rotation invariant cooccurrence matrix 19 of :20

20 Cooccurrence Matrices Cooccurrence Matrix Features (22) Retaining the entire cooccurrence matrix is not efficient. typically, the widely spread Haralick features are extracted for later classification: Energy: Entropy: Contrast: Homogeneity: Max probability: Energy and entropy are both measures of how concentrated the entries occur. Energy is high if has strong peaks, and the entropy reaches a max if all entries are equal probable. Contrast and homogeneity are more specifically oriented towards the entries being concentrated near the diagonal (then contrast is small and homogeneity is large). 20 of :20

21 Cooccurrence Matrices Haralick Feature Examples (23) Energy Entropy (a) Contrast (b) Homogeneity Max. Probability (d) (e) 21 of :20

22 Fourier Features Fourier Features (25) Frequency related concepts like a texture being periodic or coarse-grained led to features extracted from the Fourier power spectrum. If a texture is highly regular (periodic) it can be expected to have dominant peaks in its power spectrum. To evaluate how strongly peaked a spectrum is, one can calculate its entropy. The term Entropy originates from Physics as was later adapted to Information Theory and is a measure for the amount of information stored in a particular signal. It is defined as (8.4) where is the probability density function (PDF) estimated with the histogram of the powerspectrum, is usually 2 yielding the unit bit for the entropy. Fig 8.12: Sample texture Fig 8.13: Corresponding Fourier power spectrum 22 of :20

23 Fourier Features Fourier Features: Entropy Example (26) Example of the Fourier power spectra and its entropy of four different textures. Entropy: Entropy: Entropy: Entropy: Fig 8.14 Power spectrum and entropy example 23 of :20

24 Fourier Features Fourier Features (2) (27) Other commonly used Fourier features use integrated power values to assess the textures coarseness and its directionality. Integration is performed over two different kinds of regions in the Fourier domain, namely Fig 8.15: concentric rings around the origin to assess the coarseness, and sectors from the apex to assess the directionality (8.5) Fig 8.16: (8.6) 24 of :20

25 Fourier Features Disadvantage of the Fourier Features (28) As the Fourier power spectrum is typically calculated over the entire image, these features collect global information instead of localised texture properties. 25 of :20

26 Filter Banks Filter Banks (30) A simple but very effective method for extracting texture information is to convolve the image with a series of convolution filters. The output of these Filter banks are then combined into a feature vector used for classification. Fig 8.17: Principle of filter banks 26 of :20

27 Laws Filters Laws Filters (32) The Laws Filters use a fixed set of ad hoc filters of size,, and, created by the convolution of row and column filters with coefficients specified in Table 8.1. Usually, the output of these filters is non-linearly transformed and then averaged over larger regions in order to get an energy. The energies for the different filters are then combined into a feature vector, that can be used for segmentation or classification. Tab 8.1: 1D Laws Filter 27 of :20

28 Gabor Filters Gabor Filters (34) In contrast to the ad hoc chosen Law filters, the Gabor filters offer a more principled alternative with a solid mathematical background. The Gabor filters are constructed by modulating a Gaussian curve with a sine function. This yields for vertically oriented Gabor filters: (8.7) where the parameter determines the width of the Gaussian envelope, specifies the frequency of the modulating cosine, while specifies the phase. : Fig 8.18: A Gabor filter profile (in the spatial domain) with, and Filters with other orientations can be easily obtained by applying a rigid rotation of the -plane. 28 of :20

29 Gabor Filters Gabor Filters (35) Guess what the Power Spectrum of the Gabor filter looks like 29 of :20

30 Gabor Filters Gabor Filter Power Spectrum (36) The Gabor filter samples the Fourier domain as can be seen from its Fourier transform (8.8) where. The corresponding power spectrum plot is depicted in Fig The Gabor filter being a product of a Gaussian and a cosine, it's Fourier transform is known to be the convolution of a Gaussian and a pair of Dirac impulses. If we rotate the filter, the positions of these two lobes will undergo the same rotation in the frequency domain. Fig 8.19: Fourier power spectrum of the Gabor filter 30 of :20

31 Gabor Filters Gabor Filter in the Frequency Domain (37) By increasing the number of parameters, the Gaussian lobes can be made elliptical. Fig 8.20 shows how the Gaussian profile can be made elliptical to optimally tessellate the frequency domain. Gabor filters provide the best possible simultaneous localisation in the spatial and the frequency domain, i.e.. Fig 8.20: Gabor filter in the frequency domain Fig 8.21: Gabor filter in the frequency domain 31 of :20

32 Gabor Filters Gabor Filter Example (38) (a) (b) (c) Fig 8.22: (a) Input texture, (b) output of a Gabor filter tuned to detect horizontal structures, (c) same for vertical structures. 32 of :20

33 Eigenfilters Eigenfilters (40) The Laws and Gabor filter banks have fixed coefficients. If a specific texture should be analysed, it can be interesting to finetune the filter set for this texture. Basic Idea 1. Consider a small neighbourhood e.g. pixels 2. All 9 pixels in this mask are put in a 9-vector 3. When shifting the mask over the image many such samples of this 9-vector are acquired 4. Calculate the corresponding covariance matrix 5. The principal components yield a new orthogonal basis for these 9-vectors 6. Arranging the components of all eigenvectors (principal components) yields in total 9 convolution filters of size Fig 8.23: Principle of the eigenfilter design 33 of :20

34 Eigenfilters Generalisation of Eigenfilters (41) In most practical cases it is useful to also take the scale of the texture into account. This is done by extracting the periodicity of the pattern with e.g. autocorrelation and then adapting the size of the filters accordingly. This will generally result in large and therefore unpractical convolution masks. A way out is to consider sparse masks, i.e. masks where most of the elements are set to zero, see Fig Fig 8.24: Structure of the eigenfilter convolution mask. The black squares indicate positions in the convolution mask that have non-zero coefficients 34 of :20

35 Eigenfilters Eigenfilter Example (42) Task Develop a computationally efficient filter that can reliably detect such flaws. Fig 8.25: Textile sample with a flaw 35 of :20

36 Eigenfilters Eigenfilter Example - Periodicity of the Texture (43) %load the flawless image img=imread('textile.png'); img=double(img(:,:,1)); imshow(img,[]); % Autocorrelation the image % with a small patch of itself r=imfilter(img,img(1:30,1:30)); % plot the 1D graphs in x- and % y-direction plot(r(:,100)); plot(r(100,:)); % read out the periodicity dy = 7; dx = 9; As the horizontal periodicity was found to be pixels and the vertical periodicity was found as, a convolution mask of size is used. To simplify the calculations we try a sparse convolution mask with the arrangement as in Fig Fig 8.26: Correct textile sample Fig 8.27: Autocorrelation result in x-direction Fig 8.28: Autocorrelation result in y-direction 36 of :20

37 Eigenfilters Eigenfilter Example - Extract the Sample Vectors (44) In the second step we shift the sparse mask over the entire image of the flawless textile image and extract the 9-element vectors at each location. These sample vectors are then stored in the matrix. [m n] = size(img); X=zeros((n-dx)*(m-dy),9); w=[ ; ; ; ; ; ; ]; w=find(w>0); [m n] = size(img); pos=1; for x=1:n-dx for y=1:m-dy tmp=img(y:y+dy-1,x:x+dx-1); X(pos,:)=tmp(w)'; pos=pos+1; end end Fig 8.29: Extraction of the sample vectors 37 of :20

38 Eigenfilters Eigenfilter Example - Calculation of the Eigenfilters (45) The Eigenfilters can the easily calculated with the MATLAB function princomp. [pc,score,lambda] = princomp(x,'econ'); Fig 8.30: Construction of the eigenfilters 38 of :20

39 Eigenfilters Eigenfilter Example - Eigenfilter the Flawed Textile (46) In this step we filter the flawed image with all the eigenfilters we previously found. flawed = imread('textile_flaw.png'); flawed = double(flawed(:,:,1)); flawed_flt = zeros(m,n,9); for i = 1:9 kern = zeros(7,9); kern(w) = pc(:,i); flawed_flt(:,:,i) = imfilter(flawed,kern); figure, imshow(flawed_flt(:,:,i), []); end It can be clearly seen, that the flaw shows up in several of the eigenfiltred images. Filtered with PC 1 Filtered with PC 2 Filtered with PC 3 Filtered with PC 4 Filtered with PC 5 Filtered with PC 6 Filtered with PC 7 Filtered with PC 8 Filtered with PC 9 39 of :20

40 Eigenfilters Eigenfilter Example - Smooth the Filter Example (47) The aim of this step is to generate images with more homogeneous intensities. This is achieved by squaring all values and boxfiltering the image with a kernel of size. flawed_pwr = zeros(m,n,9); for i = 1:9 flawed_pwr(:,:,i) = imfilter(flawed_flt(:,:,i).^2,ones(5)); figure, imshow(flawed_pwr(:,:,i),[]); end Local energy filtered with PC 1 Local energy filtered with PC 2 Local energy filtered with PC 3 Local energy filtered with PC 5 Local energy filtered with PC 6 Local energy filtered with PC 7 Local energy filtered with PC 4 Local energy filtered with PC 8 Local energy filtered with PC 9 40 of :20

41 Eigenfilters Eigenfilter Example - Mahalanobis Distance (48) In this last step, the energy values are combined into a single distance (Mahalanobis) to the typical energies. For this we need the energies of the flawless textile: intact_pwr = zeros(m,n,9); for i = 1:9 kern = zeros(7,9); kern(w) = pc(:,i); Fig 8.31: Mahalanobis distance for the image in Fig 8.25 tmp=imfilter(imfilter(img,kern).^2,ones(5)); figure, imshow(tmp,[]); intact_pwr(:,:,i) = tmp; end then we calculate the Mahalanobis distance of every point of the defective textile to the energy values seen in the flawless textile image. intact_pwr2 = reshape(intact_pwr,n*m,9); flawed_pwr2 = reshape(flawed_pwr,n*m,9); r = mahal(flawed_pwr2,intact_pwr2); out = reshape(r,m,n); figure, imshow(out,[]); 41 of :20

42 Texture Discrimination Texture Discrimination (50) In Texture discrimination the extracted texture measures are analysed and the image split into regions, each corresponding to a perceptually homogeneous texture. In texture segmentation we always face the trade-off between sample size and accuracy. The bigger the sample size, the better the accuracy of feature estimation; however, this allows a coarse segmentation only. Fig 8.33: Texture measures for small regions or even for individual pixels are known Fig 8.32: Texture segmentation principle: texture discrimination Fig 8.34: Regions with similar texture measures are grouped together 42 of :20

43 Region Growing Region Growing Methods Segmentation may be viewed as a process that partitions subregions,, such that (52) into (a) (b) is a connected region, (c) for all and, (d) for, and (e) for where is a logical predicate over points in the set. 43 of :20

44 Region Growing Region Growing by Pixel Aggregation (53) The simplest Region growing approach is pixel aggregation, which starts with a set of seed points and from these grows regions by appending to each seed point those neighbouring pixels that have similar properties i.e. texture measures. (a) Step 1 (b) Step 2 (c) Step 3 (d) Step 4 Fig 8.35: Region growing by pixel aggregation using an absolute difference of less than 3 between intensity levels and 4-connectivity 44 of :20

45 Other Methods Other Methods (55) Below is a list other commonly used texture discrimination methods Watershed k-means clustering Bayesian classification Artificial neural networks (ANN) Estimation Theory - Maximum likelihood Split-and-Merge 45 of :20

46 Reduction of Dimensionality Reduction of Dimensionality (57) In many cases it is desirable to prior reduce the dimensionality of the extracted texture measures. PCA is the method of choice as can be seen in Fig Fig 8.36: Successfull dimensionality reduction with PCA Fig 8.37: Failed PCA dimensionality reduction Beware: in some special cases dimensionality reduction with PCA fails, see Fig of :20

47 Texture Classification Texture Classification (59) Once the texture features have been extracted and regions with similar properties grouped, all that remains is to classify the different regions into known textures. The following methods are given as a reference but will not be discussed in this lecture Bayesian Decision Theory Artificial Neural Networks (ANN) Support Vector Machines (SVM) Fig 8.38: Texture segmentation principle: Texture classification 47 of :20

48 References References (61) Robert M Haralick, K Shanmugam, Its'hak Dinstein (1973). "Textural Features for Image Classification". IEEE Transactions on Systems, Man, and Cybernetics SMC-3 (6): of :20

Texture Analysis. Selim Aksoy Department of Computer Engineering Bilkent University

Texture Analysis. Selim Aksoy Department of Computer Engineering Bilkent University Texture Analysis Selim Aksoy Department of Computer Engineering Bilkent University saksoy@cs.bilkent.edu.tr Texture An important approach to image description is to quantify its texture content. Texture

More information

Texture. Frequency Descriptors. Frequency Descriptors. Frequency Descriptors. Frequency Descriptors. Frequency Descriptors

Texture. Frequency Descriptors. Frequency Descriptors. Frequency Descriptors. Frequency Descriptors. Frequency Descriptors Texture The most fundamental question is: How can we measure texture, i.e., how can we quantitatively distinguish between different textures? Of course it is not enough to look at the intensity of individual

More information

TEXTURE. Plan for today. Segmentation problems. What is segmentation? INF 4300 Digital Image Analysis. Why texture, and what is it?

TEXTURE. Plan for today. Segmentation problems. What is segmentation? INF 4300 Digital Image Analysis. Why texture, and what is it? INF 43 Digital Image Analysis TEXTURE Plan for today Why texture, and what is it? Statistical descriptors First order Second order Gray level co-occurrence matrices Fritz Albregtsen 8.9.21 Higher order

More information

ECE 176 Digital Image Processing Handout #14 Pamela Cosman 4/29/05 TEXTURE ANALYSIS

ECE 176 Digital Image Processing Handout #14 Pamela Cosman 4/29/05 TEXTURE ANALYSIS ECE 176 Digital Image Processing Handout #14 Pamela Cosman 4/29/ TEXTURE ANALYSIS Texture analysis is covered very briefly in Gonzalez and Woods, pages 66 671. This handout is intended to supplement that

More information

CHAPTER 4 TEXTURE FEATURE EXTRACTION

CHAPTER 4 TEXTURE FEATURE EXTRACTION 83 CHAPTER 4 TEXTURE FEATURE EXTRACTION This chapter deals with various feature extraction technique based on spatial, transform, edge and boundary, color, shape and texture features. A brief introduction

More information

Schedule for Rest of Semester

Schedule for Rest of Semester Schedule for Rest of Semester Date Lecture Topic 11/20 24 Texture 11/27 25 Review of Statistics & Linear Algebra, Eigenvectors 11/29 26 Eigenvector expansions, Pattern Recognition 12/4 27 Cameras & calibration

More information

Fundamentals of Digital Image Processing

Fundamentals of Digital Image Processing \L\.6 Gw.i Fundamentals of Digital Image Processing A Practical Approach with Examples in Matlab Chris Solomon School of Physical Sciences, University of Kent, Canterbury, UK Toby Breckon School of Engineering,

More information

Image Sampling and Quantisation

Image Sampling and Quantisation Image Sampling and Quantisation Introduction to Signal and Image Processing Prof. Dr. Philippe Cattin MIAC, University of Basel 1 of 46 22.02.2016 09:17 Contents Contents 1 Motivation 2 Sampling Introduction

More information

Image Sampling & Quantisation

Image Sampling & Quantisation Image Sampling & Quantisation Biomedical Image Analysis Prof. Dr. Philippe Cattin MIAC, University of Basel Contents 1 Motivation 2 Sampling Introduction and Motivation Sampling Example Quantisation Example

More information

5. Feature Extraction from Images

5. Feature Extraction from Images 5. Feature Extraction from Images Aim of this Chapter: Learn the Basic Feature Extraction Methods for Images Main features: Color Texture Edges Wie funktioniert ein Mustererkennungssystem Test Data x i

More information

Texture. Texture is a description of the spatial arrangement of color or intensities in an image or a selected region of an image.

Texture. Texture is a description of the spatial arrangement of color or intensities in an image or a selected region of an image. Texture Texture is a description of the spatial arrangement of color or intensities in an image or a selected region of an image. Structural approach: a set of texels in some regular or repeated pattern

More information

TEXTURE ANALYSIS USING GABOR FILTERS

TEXTURE ANALYSIS USING GABOR FILTERS TEXTURE ANALYSIS USING GABOR FILTERS Texture Types Definition of Texture Texture types Synthetic Natural Stochastic < Prev Next > Texture Definition Texture: the regular repetition of an element or pattern

More information

Region-based Segmentation

Region-based Segmentation Region-based Segmentation Image Segmentation Group similar components (such as, pixels in an image, image frames in a video) to obtain a compact representation. Applications: Finding tumors, veins, etc.

More information

CHAPTER 1 Introduction 1. CHAPTER 2 Images, Sampling and Frequency Domain Processing 37

CHAPTER 1 Introduction 1. CHAPTER 2 Images, Sampling and Frequency Domain Processing 37 Extended Contents List Preface... xi About the authors... xvii CHAPTER 1 Introduction 1 1.1 Overview... 1 1.2 Human and Computer Vision... 2 1.3 The Human Vision System... 4 1.3.1 The Eye... 5 1.3.2 The

More information

9 length of contour = no. of horizontal and vertical components + ( 2 no. of diagonal components) diameter of boundary B

9 length of contour = no. of horizontal and vertical components + ( 2 no. of diagonal components) diameter of boundary B 8. Boundary Descriptor 8.. Some Simple Descriptors length of contour : simplest descriptor - chain-coded curve 9 length of contour no. of horiontal and vertical components ( no. of diagonal components

More information

ELEC Dr Reji Mathew Electrical Engineering UNSW

ELEC Dr Reji Mathew Electrical Engineering UNSW ELEC 4622 Dr Reji Mathew Electrical Engineering UNSW Review of Motion Modelling and Estimation Introduction to Motion Modelling & Estimation Forward Motion Backward Motion Block Motion Estimation Motion

More information

Digital Image Processing. Lecture # 15 Image Segmentation & Texture

Digital Image Processing. Lecture # 15 Image Segmentation & Texture Digital Image Processing Lecture # 15 Image Segmentation & Texture 1 Image Segmentation Image Segmentation Group similar components (such as, pixels in an image, image frames in a video) Applications:

More information

SLIDING WINDOW FOR RELATIONS MAPPING

SLIDING WINDOW FOR RELATIONS MAPPING SLIDING WINDOW FOR RELATIONS MAPPING Dana Klimesova Institute of Information Theory and Automation, Prague, Czech Republic and Czech University of Agriculture, Prague klimes@utia.cas.c klimesova@pef.czu.cz

More information

Feature Extraction and Image Processing, 2 nd Edition. Contents. Preface

Feature Extraction and Image Processing, 2 nd Edition. Contents. Preface , 2 nd Edition Preface ix 1 Introduction 1 1.1 Overview 1 1.2 Human and Computer Vision 1 1.3 The Human Vision System 3 1.3.1 The Eye 4 1.3.2 The Neural System 7 1.3.3 Processing 7 1.4 Computer Vision

More information

Learning and Inferring Depth from Monocular Images. Jiyan Pan April 1, 2009

Learning and Inferring Depth from Monocular Images. Jiyan Pan April 1, 2009 Learning and Inferring Depth from Monocular Images Jiyan Pan April 1, 2009 Traditional ways of inferring depth Binocular disparity Structure from motion Defocus Given a single monocular image, how to infer

More information

Image Segmentation and Registration

Image Segmentation and Registration Image Segmentation and Registration Dr. Christine Tanner (tanner@vision.ee.ethz.ch) Computer Vision Laboratory, ETH Zürich Dr. Verena Kaynig, Machine Learning Laboratory, ETH Zürich Outline Segmentation

More information

DIGITAL IMAGE ANALYSIS. Image Classification: Object-based Classification

DIGITAL IMAGE ANALYSIS. Image Classification: Object-based Classification DIGITAL IMAGE ANALYSIS Image Classification: Object-based Classification Image classification Quantitative analysis used to automate the identification of features Spectral pattern recognition Unsupervised

More information

Classification. Vladimir Curic. Centre for Image Analysis Swedish University of Agricultural Sciences Uppsala University

Classification. Vladimir Curic. Centre for Image Analysis Swedish University of Agricultural Sciences Uppsala University Classification Vladimir Curic Centre for Image Analysis Swedish University of Agricultural Sciences Uppsala University Outline An overview on classification Basics of classification How to choose appropriate

More information

Lecture 11: Classification

Lecture 11: Classification Lecture 11: Classification 1 2009-04-28 Patrik Malm Centre for Image Analysis Swedish University of Agricultural Sciences Uppsala University 2 Reading instructions Chapters for this lecture 12.1 12.2 in

More information

CHAPTER 2 TEXTURE CLASSIFICATION METHODS GRAY LEVEL CO-OCCURRENCE MATRIX AND TEXTURE UNIT

CHAPTER 2 TEXTURE CLASSIFICATION METHODS GRAY LEVEL CO-OCCURRENCE MATRIX AND TEXTURE UNIT CHAPTER 2 TEXTURE CLASSIFICATION METHODS GRAY LEVEL CO-OCCURRENCE MATRIX AND TEXTURE UNIT 2.1 BRIEF OUTLINE The classification of digital imagery is to extract useful thematic information which is one

More information

Texture Analysis of Painted Strokes 1) Martin Lettner, Paul Kammerer, Robert Sablatnig

Texture Analysis of Painted Strokes 1) Martin Lettner, Paul Kammerer, Robert Sablatnig Texture Analysis of Painted Strokes 1) Martin Lettner, Paul Kammerer, Robert Sablatnig Vienna University of Technology, Institute of Computer Aided Automation, Pattern Recognition and Image Processing

More information

Lecture 8 Object Descriptors

Lecture 8 Object Descriptors Lecture 8 Object Descriptors Azadeh Fakhrzadeh Centre for Image Analysis Swedish University of Agricultural Sciences Uppsala University 2 Reading instructions Chapter 11.1 11.4 in G-W Azadeh Fakhrzadeh

More information

TEXTURE ANALYSIS USING GABOR FILTERS FIL

TEXTURE ANALYSIS USING GABOR FILTERS FIL TEXTURE ANALYSIS USING GABOR FILTERS Texture Types Definition of Texture Texture types Synthetic ti Natural Stochastic < Prev Next > Texture Definition Texture: the regular repetition of an element or

More information

Image Analysis Lecture Segmentation. Idar Dyrdal

Image Analysis Lecture Segmentation. Idar Dyrdal Image Analysis Lecture 9.1 - Segmentation Idar Dyrdal Segmentation Image segmentation is the process of partitioning a digital image into multiple parts The goal is to divide the image into meaningful

More information

Haralick Parameters for Texture feature Extraction

Haralick Parameters for Texture feature Extraction Haralick Parameters for Texture feature Extraction Ms. Ashwini Raut1 raut.ashu87@gmail.com Mr.Saket J. Panchbhai2 ayur.map.patel@gmail.com Ms. Ketki S. Palsodkar3 chaitanya.dhondrikar96@gmail.com Ms.Ankita

More information

High Resolution Remote Sensing Image Classification based on SVM and FCM Qin LI a, Wenxing BAO b, Xing LI c, Bin LI d

High Resolution Remote Sensing Image Classification based on SVM and FCM Qin LI a, Wenxing BAO b, Xing LI c, Bin LI d 2nd International Conference on Electrical, Computer Engineering and Electronics (ICECEE 2015) High Resolution Remote Sensing Image Classification based on SVM and FCM Qin LI a, Wenxing BAO b, Xing LI

More information

Digital Image Processing. Prof. P.K. Biswas. Department of Electronics & Electrical Communication Engineering

Digital Image Processing. Prof. P.K. Biswas. Department of Electronics & Electrical Communication Engineering Digital Image Processing Prof. P.K. Biswas Department of Electronics & Electrical Communication Engineering Indian Institute of Technology, Kharagpur Image Segmentation - III Lecture - 31 Hello, welcome

More information

Image Enhancement Techniques for Fingerprint Identification

Image Enhancement Techniques for Fingerprint Identification March 2013 1 Image Enhancement Techniques for Fingerprint Identification Pankaj Deshmukh, Siraj Pathan, Riyaz Pathan Abstract The aim of this paper is to propose a new method in fingerprint enhancement

More information

Lecture 6: Multimedia Information Retrieval Dr. Jian Zhang

Lecture 6: Multimedia Information Retrieval Dr. Jian Zhang Lecture 6: Multimedia Information Retrieval Dr. Jian Zhang NICTA & CSE UNSW COMP9314 Advanced Database S1 2007 jzhang@cse.unsw.edu.au Reference Papers and Resources Papers: Colour spaces-perceptual, historical

More information

EE795: Computer Vision and Intelligent Systems

EE795: Computer Vision and Intelligent Systems EE795: Computer Vision and Intelligent Systems Spring 2012 TTh 17:30-18:45 WRI C225 Lecture 04 130131 http://www.ee.unlv.edu/~b1morris/ecg795/ 2 Outline Review Histogram Equalization Image Filtering Linear

More information

Experimentation on the use of Chromaticity Features, Local Binary Pattern and Discrete Cosine Transform in Colour Texture Analysis

Experimentation on the use of Chromaticity Features, Local Binary Pattern and Discrete Cosine Transform in Colour Texture Analysis Experimentation on the use of Chromaticity Features, Local Binary Pattern and Discrete Cosine Transform in Colour Texture Analysis N.Padmapriya, Ovidiu Ghita, and Paul.F.Whelan Vision Systems Laboratory,

More information

Advanced Video Content Analysis and Video Compression (5LSH0), Module 4

Advanced Video Content Analysis and Video Compression (5LSH0), Module 4 Advanced Video Content Analysis and Video Compression (5LSH0), Module 4 Visual feature extraction Part I: Color and texture analysis Sveta Zinger Video Coding and Architectures Research group, TU/e ( s.zinger@tue.nl

More information

CHAPTER 4 FEATURE EXTRACTION AND SELECTION TECHNIQUES

CHAPTER 4 FEATURE EXTRACTION AND SELECTION TECHNIQUES 69 CHAPTER 4 FEATURE EXTRACTION AND SELECTION TECHNIQUES 4.1 INTRODUCTION Texture is an important characteristic for analyzing the many types of images. It can be seen in all images, from multi spectral

More information

Biometrics Technology: Image Processing & Pattern Recognition (by Dr. Dickson Tong)

Biometrics Technology: Image Processing & Pattern Recognition (by Dr. Dickson Tong) Biometrics Technology: Image Processing & Pattern Recognition (by Dr. Dickson Tong) References: [1] http://homepages.inf.ed.ac.uk/rbf/hipr2/index.htm [2] http://www.cs.wisc.edu/~dyer/cs540/notes/vision.html

More information

Face Detection for Skintone Images Using Wavelet and Texture Features

Face Detection for Skintone Images Using Wavelet and Texture Features Face Detection for Skintone Images Using Wavelet and Texture Features 1 H.C. Vijay Lakshmi, 2 S. Patil Kulkarni S.J. College of Engineering Mysore, India 1 vijisjce@yahoo.co.in, 2 pk.sudarshan@gmail.com

More information

Texture Image Segmentation using FCM

Texture Image Segmentation using FCM Proceedings of 2012 4th International Conference on Machine Learning and Computing IPCSIT vol. 25 (2012) (2012) IACSIT Press, Singapore Texture Image Segmentation using FCM Kanchan S. Deshmukh + M.G.M

More information

Practical Image and Video Processing Using MATLAB

Practical Image and Video Processing Using MATLAB Practical Image and Video Processing Using MATLAB Chapter 18 Feature extraction and representation What will we learn? What is feature extraction and why is it a critical step in most computer vision and

More information

CS 4495 Computer Vision. Linear Filtering 2: Templates, Edges. Aaron Bobick. School of Interactive Computing. Templates/Edges

CS 4495 Computer Vision. Linear Filtering 2: Templates, Edges. Aaron Bobick. School of Interactive Computing. Templates/Edges CS 4495 Computer Vision Linear Filtering 2: Templates, Edges Aaron Bobick School of Interactive Computing Last time: Convolution Convolution: Flip the filter in both dimensions (right to left, bottom to

More information

Textural Features for Image Database Retrieval

Textural Features for Image Database Retrieval Textural Features for Image Database Retrieval Selim Aksoy and Robert M. Haralick Intelligent Systems Laboratory Department of Electrical Engineering University of Washington Seattle, WA 98195-2500 {aksoy,haralick}@@isl.ee.washington.edu

More information

CS 534: Computer Vision Texture

CS 534: Computer Vision Texture CS 534: Computer Vision Texture Ahmed Elgammal Dept of Computer Science CS 534 Texture - 1 Outlines Finding templates by convolution What is Texture Co-occurrence matrices for texture Spatial Filtering

More information

Texture Analysis and Applications

Texture Analysis and Applications Texture Analysis and Applications Chaur-Chin Chen Department of Computer Science National Tsing Hua University Hsinchu 30043, Taiwan E-mail: cchen@cs.nthu.edu.tw Tel/Fax: (03) 573-1078/572-3694 Outline

More information

Digital Image Processing

Digital Image Processing Digital Image Processing Part 9: Representation and Description AASS Learning Systems Lab, Dep. Teknik Room T1209 (Fr, 11-12 o'clock) achim.lilienthal@oru.se Course Book Chapter 11 2011-05-17 Contents

More information

Image Segmentation. Shengnan Wang

Image Segmentation. Shengnan Wang Image Segmentation Shengnan Wang shengnan@cs.wisc.edu Contents I. Introduction to Segmentation II. Mean Shift Theory 1. What is Mean Shift? 2. Density Estimation Methods 3. Deriving the Mean Shift 4. Mean

More information

2%34 #5 +,,% ! # %& ()% #% +,,%. & /%0%)( 1 ! # %& % %()# +(& ,.+/ +&0%//#/ &

2%34 #5 +,,% ! # %& ()% #% +,,%. & /%0%)( 1 ! # %& % %()# +(& ,.+/ +&0%//#/ & ! # %& ()% #% +,,%. & /%0%)( 1 2%34 #5 +,,%! # %& % %()# +(&,.+/ +&0%//#/ & & Many slides in this lecture are due to other authors; they are credited on the bottom right Topics of This Lecture Problem

More information

Periodicity Extraction using Superposition of Distance Matching Function and One-dimensional Haar Wavelet Transform

Periodicity Extraction using Superposition of Distance Matching Function and One-dimensional Haar Wavelet Transform Periodicity Extraction using Superposition of Distance Matching Function and One-dimensional Haar Wavelet Transform Dr. N.U. Bhajantri Department of Computer Science & Engineering, Government Engineering

More information

INTENSITY TRANSFORMATION AND SPATIAL FILTERING

INTENSITY TRANSFORMATION AND SPATIAL FILTERING 1 INTENSITY TRANSFORMATION AND SPATIAL FILTERING Lecture 3 Image Domains 2 Spatial domain Refers to the image plane itself Image processing methods are based and directly applied to image pixels Transform

More information

Introduction to Medical Imaging (5XSA0)

Introduction to Medical Imaging (5XSA0) 1 Introduction to Medical Imaging (5XSA0) Visual feature extraction Color and texture analysis Sveta Zinger ( s.zinger@tue.nl ) Introduction (1) Features What are features? Feature a piece of information

More information

Figure 1: Workflow of object-based classification

Figure 1: Workflow of object-based classification Technical Specifications Object Analyst Object Analyst is an add-on package for Geomatica that provides tools for segmentation, classification, and feature extraction. Object Analyst includes an all-in-one

More information

Edge and Texture. CS 554 Computer Vision Pinar Duygulu Bilkent University

Edge and Texture. CS 554 Computer Vision Pinar Duygulu Bilkent University Edge and Texture CS 554 Computer Vision Pinar Duygulu Bilkent University Filters for features Previously, thinking of filtering as a way to remove or reduce noise Now, consider how filters will allow us

More information

Simultaneous surface texture classification and illumination tilt angle prediction

Simultaneous surface texture classification and illumination tilt angle prediction Simultaneous surface texture classification and illumination tilt angle prediction X. Lladó, A. Oliver, M. Petrou, J. Freixenet, and J. Martí Computer Vision and Robotics Group - IIiA. University of Girona

More information

Statistical texture classification via histograms of wavelet filtered images

Statistical texture classification via histograms of wavelet filtered images Statistical texture classification via histograms of wavelet filtered images Author Liew, Alan Wee-Chung, Jo, Jun Hyung, Chae, Tae Byong, Chun, Yong-Sik Published 2011 Conference Title Proceedings of the

More information

then assume that we are given the image of one of these textures captured by a camera at a different (longer) distance and with unknown direction of i

then assume that we are given the image of one of these textures captured by a camera at a different (longer) distance and with unknown direction of i Image Texture Prediction using Colour Photometric Stereo Xavier Lladó 1, Joan Mart 1, and Maria Petrou 2 1 Institute of Informatics and Applications, University of Girona, 1771, Girona, Spain fllado,joanmg@eia.udg.es

More information

Image and Multidimensional Signal Processing

Image and Multidimensional Signal Processing Image and Multidimensional Signal Processing Professor William Hoff Dept of Electrical Engineering &Computer Science http://inside.mines.edu/~whoff/ Representation and Description 2 Representation and

More information

Chapter 3: Intensity Transformations and Spatial Filtering

Chapter 3: Intensity Transformations and Spatial Filtering Chapter 3: Intensity Transformations and Spatial Filtering 3.1 Background 3.2 Some basic intensity transformation functions 3.3 Histogram processing 3.4 Fundamentals of spatial filtering 3.5 Smoothing

More information

EXAM SOLUTIONS. Image Processing and Computer Vision Course 2D1421 Monday, 13 th of March 2006,

EXAM SOLUTIONS. Image Processing and Computer Vision Course 2D1421 Monday, 13 th of March 2006, School of Computer Science and Communication, KTH Danica Kragic EXAM SOLUTIONS Image Processing and Computer Vision Course 2D1421 Monday, 13 th of March 2006, 14.00 19.00 Grade table 0-25 U 26-35 3 36-45

More information

Introduction to digital image classification

Introduction to digital image classification Introduction to digital image classification Dr. Norman Kerle, Wan Bakx MSc a.o. INTERNATIONAL INSTITUTE FOR GEO-INFORMATION SCIENCE AND EARTH OBSERVATION Purpose of lecture Main lecture topics Review

More information

Image Analysis - Lecture 5

Image Analysis - Lecture 5 Texture Segmentation Clustering Review Image Analysis - Lecture 5 Texture and Segmentation Magnus Oskarsson Lecture 5 Texture Segmentation Clustering Review Contents Texture Textons Filter Banks Gabor

More information

Topic 4 Image Segmentation

Topic 4 Image Segmentation Topic 4 Image Segmentation What is Segmentation? Why? Segmentation important contributing factor to the success of an automated image analysis process What is Image Analysis: Processing images to derive

More information

Texture Modeling using MRF and Parameters Estimation

Texture Modeling using MRF and Parameters Estimation Texture Modeling using MRF and Parameters Estimation Ms. H. P. Lone 1, Prof. G. R. Gidveer 2 1 Postgraduate Student E & TC Department MGM J.N.E.C,Aurangabad 2 Professor E & TC Department MGM J.N.E.C,Aurangabad

More information

CITS 4402 Computer Vision

CITS 4402 Computer Vision CITS 4402 Computer Vision A/Prof Ajmal Mian Adj/A/Prof Mehdi Ravanbakhsh, CEO at Mapizy (www.mapizy.com) and InFarm (www.infarm.io) Lecture 02 Binary Image Analysis Objectives Revision of image formation

More information

DEPARTMENT OF ELECTRONICS AND COMMUNICATION ENGINEERING DS7201 ADVANCED DIGITAL IMAGE PROCESSING II M.E (C.S) QUESTION BANK UNIT I 1. Write the differences between photopic and scotopic vision? 2. What

More information

Edge and local feature detection - 2. Importance of edge detection in computer vision

Edge and local feature detection - 2. Importance of edge detection in computer vision Edge and local feature detection Gradient based edge detection Edge detection by function fitting Second derivative edge detectors Edge linking and the construction of the chain graph Edge and local feature

More information

October 17, 2017 Basic Image Processing Algorithms 3

October 17, 2017 Basic Image Processing Algorithms 3 Lecture 4 PPKE-ITK Textures demonstrate the difference between an artificial world of objects whose surfaces are only characterized by their color and reflectivity properties to that of real world imagery

More information

CCITC Shervan Fekri-Ershad Department of Computer Science and Engineering Shiraz University Shiraz, Fars, Iran

CCITC Shervan Fekri-Ershad Department of Computer Science and Engineering Shiraz University Shiraz, Fars, Iran Innovative Texture Database Collecting Approach and Feature Extraction Method based on Combination of Gray Tone Difference Matrixes, Local Binary Patterns, and K-means Clustering Shervan Fekri-Ershad Department

More information

Chapter 11 Representation & Description

Chapter 11 Representation & Description Chain Codes Chain codes are used to represent a boundary by a connected sequence of straight-line segments of specified length and direction. The direction of each segment is coded by using a numbering

More information

[2006] IEEE. Reprinted, with permission, from [Wenjing Jia, Huaifeng Zhang, Xiangjian He, and Qiang Wu, A Comparison on Histogram Based Image

[2006] IEEE. Reprinted, with permission, from [Wenjing Jia, Huaifeng Zhang, Xiangjian He, and Qiang Wu, A Comparison on Histogram Based Image [6] IEEE. Reprinted, with permission, from [Wenjing Jia, Huaifeng Zhang, Xiangjian He, and Qiang Wu, A Comparison on Histogram Based Image Matching Methods, Video and Signal Based Surveillance, 6. AVSS

More information

Lecture 4. Digital Image Enhancement. 1. Principle of image enhancement 2. Spatial domain transformation. Histogram processing

Lecture 4. Digital Image Enhancement. 1. Principle of image enhancement 2. Spatial domain transformation. Histogram processing Lecture 4 Digital Image Enhancement 1. Principle of image enhancement 2. Spatial domain transformation Basic intensity it tranfomation ti Histogram processing Principle Objective of Enhancement Image enhancement

More information

A COMPARISON OF WAVELET-BASED AND RIDGELET- BASED TEXTURE CLASSIFICATION OF TISSUES IN COMPUTED TOMOGRAPHY

A COMPARISON OF WAVELET-BASED AND RIDGELET- BASED TEXTURE CLASSIFICATION OF TISSUES IN COMPUTED TOMOGRAPHY A COMPARISON OF WAVELET-BASED AND RIDGELET- BASED TEXTURE CLASSIFICATION OF TISSUES IN COMPUTED TOMOGRAPHY Lindsay Semler Lucia Dettori Intelligent Multimedia Processing Laboratory School of Computer Scienve,

More information

Digital Image Classification Geography 4354 Remote Sensing

Digital Image Classification Geography 4354 Remote Sensing Digital Image Classification Geography 4354 Remote Sensing Lab 11 Dr. James Campbell December 10, 2001 Group #4 Mark Dougherty Paul Bartholomew Akisha Williams Dave Trible Seth McCoy Table of Contents:

More information

Computer Vision 2. SS 18 Dr. Benjamin Guthier Professur für Bildverarbeitung. Computer Vision 2 Dr. Benjamin Guthier

Computer Vision 2. SS 18 Dr. Benjamin Guthier Professur für Bildverarbeitung. Computer Vision 2 Dr. Benjamin Guthier Computer Vision 2 SS 18 Dr. Benjamin Guthier Professur für Bildverarbeitung Computer Vision 2 Dr. Benjamin Guthier 1. IMAGE PROCESSING Computer Vision 2 Dr. Benjamin Guthier Content of this Chapter Non-linear

More information

Statistical Texture Analysis

Statistical Texture Analysis Statistical Texture Analysis G.. Srinivasan, and Shobha G. Abstract This paper presents an overview of the methodologies and algorithms for statistical texture analysis of 2D images. Methods for digital-image

More information

(Refer Slide Time: 0:51)

(Refer Slide Time: 0:51) Introduction to Remote Sensing Dr. Arun K Saraf Department of Earth Sciences Indian Institute of Technology Roorkee Lecture 16 Image Classification Techniques Hello everyone welcome to 16th lecture in

More information

Digital Image Processing. Prof. P. K. Biswas. Department of Electronic & Electrical Communication Engineering

Digital Image Processing. Prof. P. K. Biswas. Department of Electronic & Electrical Communication Engineering Digital Image Processing Prof. P. K. Biswas Department of Electronic & Electrical Communication Engineering Indian Institute of Technology, Kharagpur Lecture - 21 Image Enhancement Frequency Domain Processing

More information

Multimedia Computing: Algorithms, Systems, and Applications: Edge Detection

Multimedia Computing: Algorithms, Systems, and Applications: Edge Detection Multimedia Computing: Algorithms, Systems, and Applications: Edge Detection By Dr. Yu Cao Department of Computer Science The University of Massachusetts Lowell Lowell, MA 01854, USA Part of the slides

More information

Norbert Schuff VA Medical Center and UCSF

Norbert Schuff VA Medical Center and UCSF Norbert Schuff Medical Center and UCSF Norbert.schuff@ucsf.edu Medical Imaging Informatics N.Schuff Course # 170.03 Slide 1/67 Objective Learn the principle segmentation techniques Understand the role

More information

COMPARISION OF NORMAL Vs HERNIATED CERVICAL IMAGES USING GRAY LEVEL TEXTURE FEATURES

COMPARISION OF NORMAL Vs HERNIATED CERVICAL IMAGES USING GRAY LEVEL TEXTURE FEATURES COMPARISION OF NORMAL Vs HERNIATED CERVICAL IMAGES USING GRAY LEVEL TEXTURE FEATURES C.Malarvizhi 1 and P.Balamurugan 2 1 Ph.D Scholar, India 2 Assistant Professor,India Department Computer Science, Government

More information

EXAM SOLUTIONS. Computer Vision Course 2D1420 Thursday, 11 th of march 2003,

EXAM SOLUTIONS. Computer Vision Course 2D1420 Thursday, 11 th of march 2003, Numerical Analysis and Computer Science, KTH Danica Kragic EXAM SOLUTIONS Computer Vision Course 2D1420 Thursday, 11 th of march 2003, 8.00 13.00 Exercise 1 (5*2=10 credits) Answer at most 5 of the following

More information

Artifacts and Textured Region Detection

Artifacts and Textured Region Detection Artifacts and Textured Region Detection 1 Vishal Bangard ECE 738 - Spring 2003 I. INTRODUCTION A lot of transformations, when applied to images, lead to the development of various artifacts in them. In

More information

Points Lines Connected points X-Y Scatter. X-Y Matrix Star Plot Histogram Box Plot. Bar Group Bar Stacked H-Bar Grouped H-Bar Stacked

Points Lines Connected points X-Y Scatter. X-Y Matrix Star Plot Histogram Box Plot. Bar Group Bar Stacked H-Bar Grouped H-Bar Stacked Plotting Menu: QCExpert Plotting Module graphs offers various tools for visualization of uni- and multivariate data. Settings and options in different types of graphs allow for modifications and customizations

More information

Types of image feature and segmentation

Types of image feature and segmentation COMP3204/COMP6223: Computer Vision Types of image feature and segmentation Jonathon Hare jsh2@ecs.soton.ac.uk Image Feature Morphology Recap: Feature Extractors image goes in Feature Extractor featurevector(s)

More information

Lecture 6: Edge Detection

Lecture 6: Edge Detection #1 Lecture 6: Edge Detection Saad J Bedros sbedros@umn.edu Review From Last Lecture Options for Image Representation Introduced the concept of different representation or transformation Fourier Transform

More information

IMAGE ANALYSIS, CLASSIFICATION, and CHANGE DETECTION in REMOTE SENSING

IMAGE ANALYSIS, CLASSIFICATION, and CHANGE DETECTION in REMOTE SENSING SECOND EDITION IMAGE ANALYSIS, CLASSIFICATION, and CHANGE DETECTION in REMOTE SENSING ith Algorithms for ENVI/IDL Morton J. Canty с*' Q\ CRC Press Taylor &. Francis Group Boca Raton London New York CRC

More information

Data: a collection of numbers or facts that require further processing before they are meaningful

Data: a collection of numbers or facts that require further processing before they are meaningful Digital Image Classification Data vs. Information Data: a collection of numbers or facts that require further processing before they are meaningful Information: Derived knowledge from raw data. Something

More information

The role of Fisher information in primary data space for neighbourhood mapping

The role of Fisher information in primary data space for neighbourhood mapping The role of Fisher information in primary data space for neighbourhood mapping H. Ruiz 1, I. H. Jarman 2, J. D. Martín 3, P. J. Lisboa 1 1 - School of Computing and Mathematical Sciences - Department of

More information

Noise Model. Important Noise Probability Density Functions (Cont.) Important Noise Probability Density Functions

Noise Model. Important Noise Probability Density Functions (Cont.) Important Noise Probability Density Functions Others -- Noise Removal Techniques -- Edge Detection Techniques -- Geometric Operations -- Color Image Processing -- Color Spaces Xiaojun Qi Noise Model The principal sources of noise in digital images

More information

Segmentation and Grouping

Segmentation and Grouping Segmentation and Grouping How and what do we see? Fundamental Problems ' Focus of attention, or grouping ' What subsets of pixels do we consider as possible objects? ' All connected subsets? ' Representation

More information

Classification. Vladimir Curic. Centre for Image Analysis Swedish University of Agricultural Sciences Uppsala University

Classification. Vladimir Curic. Centre for Image Analysis Swedish University of Agricultural Sciences Uppsala University Classification Vladimir Curic Centre for Image Analysis Swedish University of Agricultural Sciences Uppsala University Outline An overview on classification Basics of classification How to choose appropriate

More information

Analysis of Irregularly Shaped Texture Regions 1

Analysis of Irregularly Shaped Texture Regions 1 Computer Vision and Image Understanding 84, 62 76 (2001) doi:10.1006/cviu.2001.0941, available online at http://www.idealibrary.com on Analysis of Irregularly Shaped Texture Regions 1 Pedro García-Sevilla

More information

What will we learn? Neighborhood processing. Convolution and correlation. Neighborhood processing. Chapter 10 Neighborhood Processing

What will we learn? Neighborhood processing. Convolution and correlation. Neighborhood processing. Chapter 10 Neighborhood Processing What will we learn? Lecture Slides ME 4060 Machine Vision and Vision-based Control Chapter 10 Neighborhood Processing By Dr. Debao Zhou 1 What is neighborhood processing and how does it differ from point

More information

Digital Image Processing

Digital Image Processing Digital Image Processing Third Edition Rafael C. Gonzalez University of Tennessee Richard E. Woods MedData Interactive PEARSON Prentice Hall Pearson Education International Contents Preface xv Acknowledgments

More information

Feature extraction. Bi-Histogram Binarization Entropy. What is texture Texture primitives. Filter banks 2D Fourier Transform Wavlet maxima points

Feature extraction. Bi-Histogram Binarization Entropy. What is texture Texture primitives. Filter banks 2D Fourier Transform Wavlet maxima points Feature extraction Bi-Histogram Binarization Entropy What is texture Texture primitives Filter banks 2D Fourier Transform Wavlet maxima points Edge detection Image gradient Mask operators Feature space

More information

TEXTURE CLASSIFICATION METHODS: A REVIEW

TEXTURE CLASSIFICATION METHODS: A REVIEW TEXTURE CLASSIFICATION METHODS: A REVIEW Ms. Sonal B. Bhandare Prof. Dr. S. M. Kamalapur M.E. Student Associate Professor Deparment of Computer Engineering, Deparment of Computer Engineering, K. K. Wagh

More information

FEATURE EXTRACTION TECHNIQUES FOR IMAGE RETRIEVAL USING HAAR AND GLCM

FEATURE EXTRACTION TECHNIQUES FOR IMAGE RETRIEVAL USING HAAR AND GLCM FEATURE EXTRACTION TECHNIQUES FOR IMAGE RETRIEVAL USING HAAR AND GLCM Neha 1, Tanvi Jain 2 1,2 Senior Research Fellow (SRF), SAM-C, Defence R & D Organization, (India) ABSTRACT Content Based Image Retrieval

More information

Lecture 4: Spatial Domain Transformations

Lecture 4: Spatial Domain Transformations # Lecture 4: Spatial Domain Transformations Saad J Bedros sbedros@umn.edu Reminder 2 nd Quiz on the manipulator Part is this Fri, April 7 205, :5 AM to :0 PM Open Book, Open Notes, Focus on the material

More information

Analysis of Functional MRI Timeseries Data Using Signal Processing Techniques

Analysis of Functional MRI Timeseries Data Using Signal Processing Techniques Analysis of Functional MRI Timeseries Data Using Signal Processing Techniques Sea Chen Department of Biomedical Engineering Advisors: Dr. Charles A. Bouman and Dr. Mark J. Lowe S. Chen Final Exam October

More information