computer and robot vision i
DESCRIPTION
Computer and Robot Vision I. Chapter 8 The Facet Model. Presented by: 蘇唯誠 [email protected] 指導教授 : 傅楸善 博士. 8.1 Introduction. facet model: image as continuum or piecewise continuous intensity surface observed digital image: noisy discretized sampling of distorted version Why? - PowerPoint PPT PresentationTRANSCRIPT
![Page 1: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/1.jpg)
Digital Camera and Computer Vision LaboratoryDepartment of Computer Science and Information Engineering
National Taiwan University, Taipei, Taiwan, R.O.C.
Computer and Robot Vision I
Chapter 8The Facet Model
Presented by: 蘇唯誠[email protected]指導教授 : 傅楸善 博士
![Page 2: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/2.jpg)
DC & CV Lab.CSIE NTU
8.1 Introduction facet model: image as continuum or piecewise
continuous intensity surface observed digital image: noisy discretized
sampling of distorted version Why?
Before processing images, must define what the conditioning means with respect to the underlying gray level intensity surface.
Conditioning estimates informative pattern.
![Page 3: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/3.jpg)
8.1 Introduction
DC & CV Lab.CSIE NTU
![Page 4: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/4.jpg)
DC & CV Lab.CSIE NTU
8.1 Introduction general forms:
1. piecewise constant (flat facet model), ideal region: constant gray level 2. piecewise linear (sloped facet model), ideal region: sloped plane gray level 3. piecewise quadratic, gray level surface: bivariate quadratic 4. piecewise cubic, gray level surface: cubic
surfaces
![Page 5: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/5.jpg)
DC & CV Lab.CSIE NTU
8.2 Relative Maxima
a simple labeling application Detect and locate all relative maxima
relative maxima: first derivative zero second derivative negative
![Page 6: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/6.jpg)
8.2 Relative Maxima
One-dimensional observation sequence ,……
We can least-squares fit a quadratic function ,
where , to each group of 2k+1 successive observations
DC & CV Lab.CSIE NTU
![Page 7: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/7.jpg)
8.2 Relative Maxima
The squared fitting error for the group of 2k+1 can be expressed by
DC & CV Lab.CSIE NTU
![Page 8: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/8.jpg)
8.2 Relative Maxima
Taking partial derivatives of
DC & CV Lab.CSIE NTU
![Page 9: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/9.jpg)
8.2 Relative Maxima
Assume k = 1, setting these partial derivatives to zero
DC & CV Lab.CSIE NTU
![Page 10: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/10.jpg)
8.2 Relative Maxima
DC & CV Lab.CSIE NTU
The quadratic has relative extrema at The extremum is a relative maximum when
![Page 11: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/11.jpg)
8.2 Relative Maxima
DC & CV Lab.CSIE NTU
![Page 12: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/12.jpg)
DC & CV Lab.CSIE NTU
8.3 Sloped Facet Parameter and Error Estimation
Least-squares procedure: to estimate sloped facet parameter, noise variance
image function
Where is a random variable indexed on , which represents noise.
and that the noise for any two pixels is independent.
),(),( crcrcrg
![Page 13: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/13.jpg)
8.3 Sloped Facet Parameter and Error Estimation
determine parameters that minimize the sum of the squared differences
DC & CV Lab.CSIE NTU
Rr Cc
crgcr2
)],(ˆˆˆ[2
![Page 14: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/14.jpg)
8.3 Sloped Facet Parameter and Error Estimation
Taking the partial derivatives of and setting them to zero
DC & CV Lab.CSIE NTU
![Page 15: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/15.jpg)
8.3 Sloped Facet Parameter and Error Estimation
Without loss of generality, we choose our coordinate RxC so that the center of the neighborhood RxC has coordinates (0, 0)
The symmetry in the chosen coordinate system leads to
DC & CV Lab.CSIE NTU
![Page 16: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/16.jpg)
8.3 Sloped Facet Parameter and Error Estimation
Solving for , , and
DC & CV Lab.CSIE NTU
![Page 17: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/17.jpg)
8.3 Sloped Facet Parameter and Error Estimation
Replacing by
DC & CV Lab.CSIE NTU
![Page 18: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/18.jpg)
8.3 Sloped Facet Parameter and Error Estimation
is noise having mean 0 and variance
DC & CV Lab.CSIE NTU
![Page 19: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/19.jpg)
8.3 Sloped Facet Parameter and Error Estimation
Examining the squared error residual
DC & CV Lab.CSIE NTU
![Page 20: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/20.jpg)
8.3 Sloped Facet Parameter and Error Estimation
Using
We obtain
DC & CV Lab.CSIE NTU
![Page 21: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/21.jpg)
8.3 Sloped Facet Parameter and Error Estimation
chi-distribution
Assume , , …… , are independent variable , where
is distributed according to the chi-squared distribution with n degrees of freedom.
DC & CV Lab.CSIE NTU
![Page 22: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/22.jpg)
8.3 Sloped Facet Parameter and Error Estimation
is a chi-squared distribution with degrees of freedom.
DC & CV Lab.CSIE NTU
![Page 23: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/23.jpg)
8.3 Sloped Facet Parameter and Error Estimation
Because , , and are independent normal distributions
is a chi-squared distribution with 3 degrees of freedom.
DC & CV Lab.CSIE NTU
![Page 24: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/24.jpg)
8.3 Sloped Facet Parameter and Error Estimation
is distributed as a chi-squared variate with degrees of freedom.
This means that can be used as an unbiased estimator for
DC & CV Lab.CSIE NTU
![Page 25: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/25.jpg)
DC & CV Lab.CSIE NTU
8.4 Facet-Based Peak Noise Removal
peak noise pixel: gray level intensity significantly differs from neighbors
(a) peak noise pixel, (b) not
![Page 26: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/26.jpg)
8.4 Facet-Based Peak Noise Removal
Center-deleted neighborhood
Where is assumed to be independent additive Gaussian noise having mean 0 and variance
DC & CV Lab.
CSIE NTU
![Page 27: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/27.jpg)
8.4 Facet-Based Peak Noise Removal
The minimizing , , and are given by
DC & CV Lab.CSIE NTU
![Page 28: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/28.jpg)
8.4 Facet-Based Peak Noise Removal
Under the hypothesis that is not peak noise, has a Gaussian distribution with mean 0 and variance
Hence has mean 0 and variance 1
DC & CV Lab.CSIE NTU
![Page 29: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/29.jpg)
8.4 Facet-Based Peak Noise Removal
We have already obtained that has a chi-squared distribution with degrees of freedom.
Hence has a t-distribution with degrees of freedom
DC & CV Lab.CSIE NTU
![Page 30: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/30.jpg)
8.4 Facet-Based Peak Noise Removal
t-distribution (0,1), , and are independent.
is a t-distribution with n degrees of freedom, denoted as
DC & CV Lab.CSIE NTU
![Page 31: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/31.jpg)
8.4 Facet-Based Peak Noise Removal
Let be the number satisfying
is a threshold.
DC & CV Lab.CSIE NTU
![Page 32: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/32.jpg)
8.4 Facet-Based Peak Noise Removal
If , the hypothesis of the equality of and is rejected, and the output value for the center pixel is given by .
If , the hypothesis of the equality of and is not rejected, and the output value for the center pixel is given by .
DC & CV Lab.CSIE NTU
![Page 33: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/33.jpg)
8.4 Facet-Based Peak Noise Removal
The reasonable value for p is 0.05 to 0.01
DC & CV Lab.CSIE NTU
![Page 34: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/34.jpg)
8.5 Iterated Facet Model
The iterated model for ideal image assume that the spatial of the image can be partitioned into connected regions called facets, each of which satisfies certain gray level and shape constraints.
DC & CV Lab.CSIE NTU
![Page 35: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/35.jpg)
DC & CV Lab.CSIE NTU
8.5 Iterated Facet Model gray level constraint:
gray levels in each facet must be a polynomial function of row-column coordinates
shape constraint: each facet must be sufficiently smooth in shape
![Page 36: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/36.jpg)
8.5 Iterated Facet Model
Each pixel in image is contained in different blocks.
Each block fits a polynomial model.
Set the output gray value to be that gray value fitted by the block with smallest error variance.
DC & CV Lab.CSIE NTU
![Page 37: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/37.jpg)
DC & CV Lab.CSIE NTU
8.6 Gradient-Based Facet Edge Detection
gradient-based facet edge detection: high values in first partial derivative (sharp discontinuities)
![Page 38: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/38.jpg)
8.6 Gradient-Based Facet Edge Detection
When a neighborhood is fitted with the sloped facet model , , a gradient magnitude of will result.
Use the estimated gradient as the basis for edge detection.
DC & CV Lab.CSIE NTU
![Page 39: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/39.jpg)
8.6 Gradient-Based Facet Edge Detection
, are independent
DC & CV Lab.CSIE NTU
![Page 40: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/40.jpg)
8.6 Gradient-Based Facet Edge Detection
It follows that to test the hypothesis of no edge under the assumption that , we use the statistic G:
DC & CV Lab.CSIE NTU
![Page 41: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/41.jpg)
8.6 Gradient-Based Facet Edge Detection
The false-alarm rate is the conditional probability that the edge detector classifies a pixel as an edge given that the pixel is not a edge.
Suppose the false-alarm rate is to be held to 1%. Then since , the threshold we used must be at least 9.21
DC & CV Lab.CSIE NTU
![Page 42: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/42.jpg)
DC & CV Lab.CSIE NTU
8.7 Bayesian Approach to Gradient Edge Detection
The Bayesian approach to the decision of whether or not an observed gradient magnitude G is statistically significant and therefore participates in some edge is to decide there is an edge (statistically significant gradient) when,
: given gradient magnitude conditional probability of edge
: given gradient magnitude conditional probability of nonedge
![Page 43: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/43.jpg)
8.7 Bayesian Approach to Gradient Edge Detection
Hence a decision for edge is made whenever
is known to be the density function of a variate.
DC & CV Lab.CSIE NTU
![Page 44: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/44.jpg)
8.7 Bayesian Approach to Gradient Edge Detection
Infer from
DC & CV Lab.CSIE NTU
![Page 45: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/45.jpg)
8.7 Bayesian Approach to Gradient Edge Detection
The threshold t for G is determined as that value t for which
Implies that the threshold t must satisfy that
is a user-specified prior probability of nonedge, 0.9 to 0.95 are reasonable.
DC & CV Lab.CSIE NTU
![Page 46: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/46.jpg)
DC & CV Lab.CSIE NTU
8.8 Zero-Crossing Edge Detector compare
gradient edge detector: looks for high values of first derivatives
zero-crossing edge detector: looks for relative maxima in first derivative
zero-crossing: pixel as edge if zero crossing of second directional derivative underlying gray level intensity function f takes the form
![Page 47: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/47.jpg)
DC & CV Lab.CSIE NTU
8.8.1 Discrete Orthogonal Polynomials
The underlying functions from which the directional derivative are computed are easy to represent as linear combinations of the polynomials in any polynomial basis set.
basis set: the discrete orthogonal polynomials
![Page 48: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/48.jpg)
DC & CV Lab.CSIE NTU
8.8.1 Discrete Orthogonal Polynomials
discrete orthogonal polynomial basis set of size N: polynomials deg. 0..N - 1
discrete Chebyshev polynomials: these unique polynomials
![Page 49: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/49.jpg)
DC & CV Lab.CSIE NTU
8.8.1 Discrete Orthogonal Polynomials (cont’)
discrete orthogonal polynomials can be recursively generated
,
![Page 50: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/50.jpg)
DC & CV Lab.CSIE NTU
8.8.2 Two-Dimensional Discrete Orthogonal Polynomials
2-D discrete orthogonal polynomials creatable from tensor products of 1D from above equations
r2731r4_r
![Page 51: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/51.jpg)
DC & CV Lab.CSIE NTU
the exact fitting problem is to determine such that
is minimized the result is ,for m=0,…K
for each index r, the data value d(r) is multiplied by the weight
8.8.3 Equal-Weighted Least-Squares Fitting Problem
![Page 52: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/52.jpg)
DC & CV Lab.CSIE NTU
8.8.3 Equal-Weighted Least-Squares Fitting Problem
weight
![Page 53: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/53.jpg)
DC & CV Lab.CSIE NTU
8.8.3 Equal-Weighted Least-Squares Fitting Problem (cont’)
![Page 54: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/54.jpg)
DC & CV Lab.CSIE NTU
8.8.4 Directional Derivative Edge Finder
We define the directional derivative edge finder as the operator that places an edge in all pixels having a negatively sloped zero crossing of the second directional derivative taken in the direction of the gradient
r: row c: column radius in polar coordinate angle in polar coordinate, clockwise from
column axis
![Page 55: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/55.jpg)
8.8.4 Directional Derivative Edge Finder (cont’)
DC & CV Lab.CSIE NTU
![Page 56: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/56.jpg)
DC & CV Lab.CSIE NTU
![Page 57: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/57.jpg)
DC & CV Lab.CSIE NTU
8.8.4 Directional Derivative Edge Finder (cont’)
directional derivative of f at point (r, c) in direction :
![Page 58: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/58.jpg)
DC & CV Lab.CSIE NTU
8.8.4 Directional Derivative Edge Finder (cont’)
second directional derivative of f at point (r, c) in direction :
![Page 59: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/59.jpg)
8.8.4 Directional Derivative Edge Finder (cont’)
DC & CV Lab.CSIE NTU
![Page 60: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/60.jpg)
8.8.4 Directional Derivative Edge Finder (cont’)
An edge pixel
DC & CV Lab.CSIE NTU
![Page 61: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/61.jpg)
DC & CV Lab.CSIE NTU
8.9 Integrated Directional Derivative Gradient Operator
integrated directional derivative gradient operator an operator that measures the integrated directional
derivative strength as the integral of first directional derivative taken over a square area
more accurate step edge direction
![Page 62: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/62.jpg)
DC & CV Lab.CSIE NTU
8.10 Corner Detection purpose
corners: to detect buildings in aerial images corner points: to determine displacement vectors from
image pair corners with two conditions
the occurrence of an edge significant changes in edge direction
gray scale corner detectors: detect corners directly by gray scale image
![Page 63: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/63.jpg)
DC & CV Lab.CSIE NTU
Aerial Images
![Page 64: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/64.jpg)
DC & CV Lab.CSIE NTU
立體視覺圖
![Page 65: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/65.jpg)
8.10 Corner Detection
Let be the gradient direction at coordinate
DC & CV Lab.CSIE NTU
![Page 66: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/66.jpg)
DC & CV Lab.CSIE NTU
8.11 Isotropic Derivative Magnitudes
gradient edge: from first-order isotropic derivative magnitude
determine those linear combinations of squared partial derivative of the two-dimensional functions that are invariant under rotation
rotate the coordinate system by and call the resulting function g in the new (r’,c’) coordinate
ckrkkcrf 321),(
cos'sin',sin'cos' crccrr
),()','( crfcrg
![Page 67: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/67.jpg)
DC & CV Lab.CSIE NTU
8.11 Isotropic Derivative Magnitudes
The sum of the squares of the first partials is the same constant for the original function or for the rotated function.
2223
22 )],([)],([ cr
cfcr
rfkk
232
232
22 )cossin()sincos()]','([)]','([ kkkkcrcgcr
rg
![Page 68: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/68.jpg)
DC & CV Lab.CSIE NTU
8.12 Ridges and Ravines on Digital Images
![Page 69: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/69.jpg)
DC & CV Lab.CSIE NTU
8.12 Ridges and Ravines on Digital Images
A digital ridge (ravine) occurs on a digital image when there is a simply connected sequence of pixels with gray level intensity values that are significantly higher (lower) in the sequence than those neighboring the sequence.
The facet model can be used to help accomplish ridge and ravine identification.
![Page 70: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/70.jpg)
DC & CV Lab.CSIE NTU
8.13 Topographic Primal Sketch8.13.1 Introduction
The basis of the topographic primal sketch consists of the labeling and grouping of the underlying Image-intensity surface patches according to the categories defined by monotonic, gray level, and invariant functions of directional derivatives
categories:
![Page 71: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/71.jpg)
DC & CV Lab.CSIE NTU
8.13.1 Introduction (cont’)Invariance Requirement
histogram normalization, equal probability quantization: nonlinear enhancing
For example, edges based on zero crossings of second derivatives will change in position as the monotonic gray level transformation changes Convexity of a gray level intensity surface is not
preserved under such transformation. peak, pit, ridge, valley, saddle, flat, hillside: have
required invariance
![Page 72: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/72.jpg)
DC & CV Lab.CSIE NTU
primal sketch: rich description of gray level changes present in image
description: includes type, position, orientation, fuzziness of edge
topographic primal sketch: we concentrate on all types of two-dimensional gray level variations
8.13.1 Introduction (cont’)Background
![Page 73: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/73.jpg)
DC & CV Lab.CSIE NTU
8.13.2 Peak
![Page 74: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/74.jpg)
8.13.2 Peak
The eigenvalues of the Hessian are the values of the extrema of the second directional derivative, and their associated eigenvectors are the directions in which the second directional derivative is extremized.
DC & CV Lab.CSIE NTU
![Page 75: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/75.jpg)
DC & CV Lab.CSIE NTU
8.13.2 Peak Peak (knob): local maximum in all directions peak: curvature downward in all directions at peak: gradient zero at peak: second directional derivative negative in
all directions point classified as peak if
: gradient magnitude
![Page 76: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/76.jpg)
DC & CV Lab.CSIE NTU
8.13.2 Peak : second directional derivative in direction : second directional derivative in direction
![Page 77: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/77.jpg)
DC & CV Lab.CSIE NTU
8.13.2 Pit pit (sink: bowl): local minimum in all directions pit: gradient zero, second directional derivative
positive
![Page 78: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/78.jpg)
DC & CV Lab.CSIE NTU
ridge: occurs on ridge line ridge line: a curve consisting of a series of ridge
points walk along ridge line: points to the right and left are
lower ridge line: may be flat, sloped upward, sloped
downward, curved upward… ridge: local maximum in one direction
8.13.2 Ridge
![Page 79: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/79.jpg)
DC & CV Lab.CSIE NTU
8.13.2 Ridge
![Page 80: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/80.jpg)
DC & CV Lab.CSIE NTU
8.13.2 Ravine ravine: valley: local minimum in one direction walk along ravine line: points to the right and left
are higher
![Page 81: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/81.jpg)
DC & CV Lab.CSIE NTU
8.13.2 Saddle saddle: local maximum in one direction, local
minimum in perpendicular dir. saddle: positive curvature in one direction,
negative in perpendicular dir. saddle: gradient magnitude zero saddle: extrema of second directional derivative
have opposite signs
![Page 82: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/82.jpg)
DC & CV Lab.CSIE NTU
8.13.2 Flat flat: plain: simple, horizontal
surface flat: zero gradient, no
curvature
flat: foot or shoulder or not qualified at all
foot: flat begins to turn up into a hill
shoulder: flat ending and turning down into a hill
![Page 83: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/83.jpg)
DC & CV Lab.CSIE NTU
8.13.2 Hillside hillside point: anything not covered by previous
categories hillside: nonzero gradient, no strict extrema Slope: tilted flat (constant gradient)
convex hill: curvature positive (upward)
![Page 84: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/84.jpg)
DC & CV Lab.CSIE NTU
8.13.2 Hillside concave hill: curvature negative (downward)
saddle hill: up in one direction, down in perpendicular direction
inflection point: zero crossing of second directional derivative
![Page 85: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/85.jpg)
DC & CV Lab.CSIE NTU
mathematical properties of topographic structures on continuous surfaces
8.13.2 Summary of the Topographic Categories
![Page 86: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/86.jpg)
DC & CV Lab.CSIE NTU
8.13.2 Invariance of the Topographic Categories
1. Topographic labels (peak, pit, ridge, ravine, saddle, flat, and hillside)
2. Gradient direction3. Directions of second directional derivative extrema
for peak, pit, ridge, ravine, and saddle 1.2.3. are all invariant under monotonically increasing
gray level transformations
monotonically increasing: positive derivative everywhere
![Page 87: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/87.jpg)
8.13.2 Invariance of the Topographic Categories
Let denote the transformed image :
DC & CV Lab.CSIE NTU
![Page 88: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/88.jpg)
8.13.2 Invariance of the Topographic Categories
The direction that maximizes also maximizes , thereby showing that the gradient directions are the same.
and and
DC & CV Lab.CSIE NTU
![Page 89: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/89.jpg)
DC & CV Lab.CSIE NTU
8.13.2 Ridge and Ravine Continua
entire areas of surface: may be classified as all ridge or all ravine
![Page 90: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/90.jpg)
DC & CV Lab.CSIE NTU
8.13.3 Topographic Classification Algorithm
peak, pit, ridge, ravine, saddle: likely not to occur at pixel center
peak, pit, ridge, ravine, saddle: if within pixel area, carry the label
![Page 91: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/91.jpg)
DC & CV Lab.CSIE NTU
8.13.3 Case One: No Zero Crossing
no zero crossing along either of two directions: flat or hillside
if gradient zero, then flat if gradient nonzero, then hillside Hillside: possibly inflection point, slope, convex
hill, concave hill,…
![Page 92: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/92.jpg)
DC & CV Lab.CSIE NTU
8.13.3 Case Two: One Zero Crossing
one zero crossing: peak, pit, ridge, ravine, or saddle
![Page 93: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/93.jpg)
DC & CV Lab.CSIE NTU
8.13.3 Case Three: Two Zero Crossings
LABEL1, LABEL2: assign label to each zero crossing
![Page 94: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/94.jpg)
DC & CV Lab.CSIE NTU
8.13.3 Case Four: More Then Two Zero Crossings
more than two zero crossings: choose the one closest to pixel center
more than two zero crossings: after ignoring the other, same as case 3
![Page 95: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/95.jpg)
DC & CV Lab.CSIE NTU
8.13.4 Summary of Topographic Classification Scheme
one pass through the image, at each pixel1. calculate fitting coefficients, through of cubic
polynomial2. use above coefficients to find gradient, gradient
magnitude, and the eigenvalues and eigenvector of Hessian
3. search in eigenvector direction for zero crossing of first derivative
4. recompute gradient, gradient magnitude, second derivative at each zero crossing of the first directional derivative, then classify
![Page 96: Computer and Robot Vision I](https://reader036.vdocuments.us/reader036/viewer/2022081507/568164e2550346895dd744fa/html5/thumbnails/96.jpg)
DC & CV Lab.CSIE NTU
END