Sheng-Fang Huang Chapter 11 part I.  After the image is segmented into regions, how to represent and describe these regions? ◦ In terms of its external.

Slides:



Advertisements
Similar presentations
Representation and Description
Advertisements

電腦視覺 Computer and Robot Vision I
November 12, 2013Computer Vision Lecture 12: Texture 1Signature Another popular method of representing shape is called the signature. In order to compute.
Document Image Processing
Extended Gaussian Images
WFM 6202: Remote Sensing and GIS in Water Management © Dr. Akm Saiful IslamDr. Akm Saiful Islam WFM 6202: Remote Sensing and GIS in Water Management Akm.
Image Segmentation Image segmentation (segmentace obrazu) –division or separation of the image into segments (connected regions) of similar properties.
Computer Vision Lecture 16: Region Representation
Each pixel is 0 or 1, background or foreground Image processing to
3D Skeletons Using Graphics Hardware Jonathan Bilodeau Chris Niski.
November 4, 2014Computer Vision Lecture 15: Shape Representation II 1Signature Another popular method of representing shape is called the signature. In.
EE 7730 Image Segmentation.
Regionalized Variables take on values according to spatial location. Given: Where: A “structural” coarse scale forcing or trend A random” Local spatial.
Lectures 10&11: Representation and description
Information that lets you recognise a region.
Chapter 11 Representation and Description. Preview Representing a region involves two choices: In terms of its external characteristics (its boundary)
E.G.M. PetrakisBinary Image Processing1 Binary Image Analysis Segmentation produces homogenous regions –each region has uniform gray-level –each region.
DIGITAL IMAGE PROCESSING
Introduction --Classification Shape ContourRegion Structural Syntactic Graph Tree Model-driven Data-driven Perimeter Compactness Eccentricity.
Digital Image Processing
Lecture 19 Representation and description II
Machine Vision for Robots
8D040 Basis beeldverwerking Feature Extraction Anna Vilanova i Bartrolí Biomedical Image Analysis Group bmia.bmt.tue.nl.
October 14, 2014Computer Vision Lecture 11: Image Segmentation I 1Contours How should we represent contours? A good contour representation should meet.
Analysis of shape Biomedical Image processing course, Yevhen Hlushchuk and Jukka Parviainen.
OBJECT RECOGNITION. The next step in Robot Vision is the Object Recognition. This problem is accomplished using the extracted feature information. The.
Digital Image Processing, 2nd ed. © 2002 R. C. Gonzalez & R. E. Woods Chapter 11 Representation & Description Chapter 11 Representation.
Digital Image Processing Lecture 20: Representation & Description
Lecture 5. Morphological Image Processing. 10/6/20152 Introduction ► ► Morphology: a branch of biology that deals with the form and structure of animals.
Recap CSC508.
Image Processing and Analysis (ImagePandA) 9 – Shape Christoph Lampert / Chris Wojtan.
Chapter 9.  Mathematical morphology: ◦ A useful tool for extracting image components in the representation of region shape.  Boundaries, skeletons,
8D040 Basis beeldverwerking Feature Extraction Anna Vilanova i Bartrolí Biomedical Image Analysis Group bmia.bmt.tue.nl.
ENT 273 Object Recognition and Feature Detection Hema C.R.
Intelligent Vision Systems ENT 496 Object Shape Identification and Representation Hema C.R. Lecture 7.
Digital Image Processing CSC331
Normal distribution and intro to continuous probability density functions...
Digital Image Processing CCS331 Relationships of Pixel 1.
Digital Camera and Computer Vision Laboratory Department of Computer Science and Information Engineering National Taiwan University, Taipei, Taiwan, R.O.C.
Introduction Image geometry studies rotation, translation, scaling, distortion, etc. Image topology studies, e.g., (i) the number of occurrences.
Digital Image Processing, 2nd ed. © 2002 R. C. Gonzalez & R. E. Woods Representation & Description.
Representation and Description Selim Aksoy Department of Computer Engineering Bilkent University
Introduction --Classification Shape ContourRegion Structural Syntactic Graph Tree Model-driven Data-driven Perimeter Compactness Eccentricity.
Paradigms for Graph Drawing Graph Drawing: Algorithms for the Visualization of Graphs - Chapter 2 Presented by Liana Diesendruck.
主講人 : 張緯德 1.  Image segmentation ◦ ex: edge-based, region-based  Image representation ◦ ex: Chain code, polygonal approximation signatures, skeletons.
CS654: Digital Image Analysis Lecture 36: Feature Extraction and Analysis.
Course 8 Contours. Def: edge list ---- ordered set of edge point or fragments. Def: contour ---- an edge list or expression that is used to represent.
Image Registration Advanced DIP Project
CDS 301 Fall, 2008 Image Visualization Chap. 9 November 11, 2008 Jie Zhang Copyright ©
1 Overview representing region in 2 ways in terms of its external characteristics (its boundary)  focus on shape characteristics in terms of its internal.
Digital Image Processing, 3rd ed. © 1992–2008 R. C. Gonzalez & R. E. Woods Gonzalez & Woods Chapter 11 Representation and.
Course 3 Binary Image Binary Images have only two gray levels: “1” and “0”, i.e., black / white. —— save memory —— fast processing —— many features of.
Chapter 6 Skeleton & Morphological Operation. Image Processing for Pattern Recognition Feature Extraction Acquisition Preprocessing Classification Post.
 After an image has been segmented into regions by methods such as those discussed in image segmentation chapter, the segmented pixels usually are represented.
Relationship between pixels Neighbors of a pixel – 4-neighbors (N,S,W,E pixels) == N 4 (p). A pixel p at coordinates (x,y) has four horizontal and vertical.
Materi 09 Analisis Citra dan Visi Komputer Representasi and Deskripsi 1.
Image Representation and Description – Representation Schemes
Medical Image Analysis
图像处理技术讲座(3) Digital Image Processing (3) Basic Image Operations
IT472: Digital Image Processing
Digital Image Processing Lecture 20: Representation & Description
Materi 10 Analisis Citra dan Visi Komputer
Representation and Description
Mean Shift Segmentation
Slope and Curvature Density Functions
Fitting Curve Models to Edges
Binary Image processing بهمن 92
Representation and Description
Representation and Description
Digital Image Processing, 3rd ed.
Presentation transcript:

Sheng-Fang Huang Chapter 11 part I

 After the image is segmented into regions, how to represent and describe these regions? ◦ In terms of its external characteristics (boundary) ◦ In terms of its internal characteristics (pixels in the region)

 Chain codes are used to represent a boundary as a connected sequence of straight line segments of specified length and direction. ◦ Based on 4- or 8- connectivity.  Chain code is generated by following a boundary in clockwise direction and assigning a direction to the segments connecting every pair of pixels.  Disadvantages: 1)The chain code is quite long. 2)Any small disturbance along the boundary due to noise cause change in the code. →resample the boundary with a larger grid spacing.

 The chain code of a boundary depends on the starting point and is not rotation invariant. ◦ We can normalize chain codes by using the first difference of the chain code. ◦ Example:  The 4-direction chain code:  The first difference of the code:  3: (1-2) mod 4 = 3  3: (0-1) mod 4 = 3  1: (1-0) mod 4 = 1 ……

 1-D functional representation of a boundary. Plot the distance from the centroid to the boundary as a function of angles, r(  ). ◦ Invariant to translation, but depend on the rotation and scaling.  Normalize with respect to rotation:  Select the starting point as the point farthest to the centroid.  Normalize with respect to size:  Scale all r(  ) so that they always span the same range, say, [0, 1].  Divided by the maximum.  Divided by the variance of r(  ).

 An important approach to representing the structural shape is to reduce it to a graph. ◦ This reduction may be achieved by obtaining its skeleton via a thinning algorithm (skeletonization).  Medial axis transformation (MAT): ◦ proposed by Blum [1967] 1)For each point p in region R, we find its closest neighbor in the border B. 2)If p has more than one such neighbor, it is said to belong to the medial axis (skeleton) of R.

 Thinning algorithm: iteratively delete the edge points of a region subject to the following constraints: 1)Does not remove the end points, 2)Does not break connectivity, and 3)Does not cause excessive erosion of the region.

 What is the difference between representation and description?  Simple descriptors 1)Length 2)Diameter: Diam(B)=max[D(p i, p j )] where p i and p j are two furthest points on the boundary. –The line between these two extremes are called the major axis. –The minor axis is the one perpendicular to the major axis where the major and minor axes form the basic rectangle. 3)Eccentricity = major axis/minor axis 4)Curvature: changes of slope. –Difficult in digit boundaries, because they are usually ragged. –Can be improved by using the curvature between adjacent boundary segments.

 The shape of boundary segments can be described by simple statistical moments, such as mean, variance, and higher-order moments.  The nth moment of v about its mean m is

 Figure 11.5 represented as 1-D function g(r). ◦ Treat the amplitude of g as a discrete random variable v and form an amplitude histogram p(v i ), i=0,1,…A-1, where A is the number of discrete amplitude increments in which we divide the amplitude scale.

The two endpoints are linked by a line segment. The coordinates of points are rotated until the line is horizontal. The distance of each point to the horizontal axis is normalized and treated as a discrete random variable to form an histogram p(v i ) where i=0,1,…A-1, g(r i ) is treated as the probability of which the value ri occurs.

 The moments are where ◦ The second moment measures the spread of the curve about the mean value of r ◦ The third moment measures its symmetry with reference to the mean.

 The area of a region is the number of pixels in the region.  Perimeter is the length of the boundary.  Compactness = perimeter 2 / area.

 Topology is the study of properties of a figure that are unaffected by any deformation (rubber-sheet distortion).  The number of holes: H  The number of connected components: C  Euler number E: E = C - H.  Regions represented by straight-line segments (polygonal networks), such as fig , has the following relationship in topology as  E = V - Q + F = C- H where V is the number of vertices and Q is the number of edges.