H. Kälviäinen, IPCV 2002, July 22 - August 2, 2002 , Koblenz, Germany 1 LAPPEENRANTA UNIVERSITY OF TECHNOLOGY THE DEPARTMENT OF INFORMATION TECHNOLOGY Intensive Program on Computer Vision IPCV 2002 July 22 – August 2, 2002 Koblenz, Germany http://www.uni-koblenz.de/~lb/lb_activities/ipc v02/ipcv02.html
38
Embed
Feature Extraction for Classification : Hough Transform and Gabor Filtering Heikki Kälviäinen
Intensive Program on Computer Vision IPCV 200 2 July 22 – August 2 , 200 2 Koblenz, Germany http://www.uni-koblenz.de/~lb/lb_activities/ipcv02/ipcv02.html. Feature Extraction for Classification : Hough Transform and Gabor Filtering Heikki Kälviäinen. Professor Computer Science - PowerPoint PPT Presentation
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
H. Kälviäinen, IPCV 2002, July 22 - August 2, 2002, Koblenz, Germany
1
LAPPEENRANTA UNIVERSITY OF TECHNOLOGYTHE DEPARTMENT OF INFORMATION TECHNOLOGY
H. Kälviäinen, IPCV 2002, July 22 - August 2, 2002, Koblenz, Germany
3
LAPPEENRANTA UNIVERSITY OF TECHNOLOGYTHE DEPARTMENT OF INFORMATION TECHNOLOGY
Lappeenranta University of Lappeenranta University of TechnologTechnology, Finlandy, Finland
H. Kälviäinen, IPCV 2002, July 22 - August 2, 2002, Koblenz, Germany
4
LAPPEENRANTA UNIVERSITY OF TECHNOLOGYTHE DEPARTMENT OF INFORMATION TECHNOLOGY
L ondonBer lin
Mosc ow
S t.Petersburg
Tall inn
Lappeenranta
Os lo
S tockholm
Hels ink i
H. Kälviäinen, IPCV 2002, July 22 - August 2, 2002, Koblenz, Germany
5
LAPPEENRANTA UNIVERSITY OF TECHNOLOGYTHE DEPARTMENT OF INFORMATION TECHNOLOGY
Contents
• Fundamentals of computer vision– Digital image processing – Pattern recognition & Machine vision– Fundamental steps in image processing – Applications
• Feature Extraction for Classification– Hough Transform– Gabor Filtering
H. Kälviäinen, IPCV 2002, July 22 - August 2, 2002, Koblenz, Germany
6
LAPPEENRANTA UNIVERSITY OF TECHNOLOGYTHE DEPARTMENT OF INFORMATION TECHNOLOGY
Digital Image Processing
• R. C. Gonzalez & R.E. Woods, Digital Image Processing, Addison-Wesley, 1993 : “A digital image is an image f(x,y) that has been discretized both in spatial coordinates and brightness”
• f(x,y) is a 2D intensity function where x and y are spatial coordinates and the value of f at any point (x,y) is proportional to the brightness of the image at the point
H. Kälviäinen, IPCV 2002, July 22 - August 2, 2002, Koblenz, Germany
7
LAPPEENRANTA UNIVERSITY OF TECHNOLOGYTHE DEPARTMENT OF INFORMATION TECHNOLOGY
Digital Image Processing
• A digital image consists of pixels (also called image elements, picture elements)
• For example: an image of a 256 x 256 array with 256 gray-levels (8 bits: 0 black, 255 white)– Binary images: only two values– Gray-level images: e.g. 256 values– Color images: three color components (e.g. RGB)– Spectral images: several components
8
LAPPEENRANTA UNIVERSITY OF TECHNOLOGYTHE DEPARTMENT OF INFORMATION TECHNOLOGY
H. Kälviäinen, IPCV 2002, July 22 - August 2, 2002, Koblenz, Germany
9
LAPPEENRANTA UNIVERSITY OF TECHNOLOGYTHE DEPARTMENT OF INFORMATION TECHNOLOGY
Pattern Recognition and Machine Vision
• A digital image is just a set of pixels ?• Pattern recognition = measurements and
observations from natural scenes and their automatic analysis and recognition
• Computer vision = image analysis using pattern recognition techniques
– Systems: for example, RTS Group (www.rts-group.com)
H. Kälviäinen, IPCV 2002, July 22 - August 2, 2002, Koblenz, Germany
25
LAPPEENRANTA UNIVERSITY OF TECHNOLOGYTHE DEPARTMENT OF INFORMATION TECHNOLOGY
Hough Transform• A method for global feature extraction:
– y = a x + b => b = -x a + y.– For each pixel (x,y) compute a curve b = -x a + b into the parameter space.– Alternatively the normal presentation of a line:
• Hough Transform detects sets of pixels which represent geometric primitives in a binary image.
• Lines, circles, ellipses, arbitrary shapes, etc.
• Tolerant to noise and distortions in an image, but traditional versions suffer from problems with time and space complexities.
• New variants: probabilistic and deterministic Hough Transforms.
sincos yx
26
LAPPEENRANTA UNIVERSITY OF TECHNOLOGYTHE DEPARTMENT OF INFORMATION TECHNOLOGY Hough
Transform(SHT)
H. Kälviäinen, IPCV 2002, July 22 - August 2, 2002, Koblenz, Germany
27
LAPPEENRANTA UNIVERSITY OF TECHNOLOGYTHE DEPARTMENT OF INFORMATION TECHNOLOGY
The Kernel of the Hough Transform
1. Create the set D of all edge points in a binary picture.
2. Transform each point in the set D into a parameterized curve in the parameter space.
3. Increment the cells in the parameter space determined by the parametric curve.
4. Detect local maxima in the accumulator array. Each local maximum may correspond to a parametric curve in the image space.
5. Extract the curve segments using the knowledge of the maximum positions.
H. Kälviäinen, IPCV 2002, July 22 - August 2, 2002, Koblenz, Germany
28
LAPPEENRANTA UNIVERSITY OF TECHNOLOGYTHE DEPARTMENT OF INFORMATION TECHNOLOGY
Randomized Hough Transform (RHT)• Developed in Lappeenranta Universityof Technology (LUT),FINLAND.• Xu, L., Oja, E., Kultanen, P, ”A New Curve Detection Method: Randomized Hough Transform (RHT), Pattern Recognition Letters, vol. 11, no. 5., 1990, pp. 331-338.
H. Kälviäinen, IPCV 2002, July 22 - August 2, 2002, Koblenz, Germany
29
LAPPEENRANTA UNIVERSITY OF TECHNOLOGYTHE DEPARTMENT OF INFORMATION TECHNOLOGY
The Kernel of the Randomized Hough Transform (RHT)
1. Create the set D of all edge points in a binary edge picture.
2. Select a point pair (d_i, d_j) randomly from the set D.
3. If the points do not satisfy the predefined distance limits, go to Step 2; otherwise continue to Step 4.
4. Solve the parameter space point (a, b) using the curve equation with the points (d_i, d_j).
5. Accumulate the cell A(a, b) in the accumulator space.
6. If the A(a, b) is equal to the threshold t, the parameters a and b describe the parameters of the detected curve; otherwise continue to Step 2.
H. Kälviäinen, IPCV 2002, July 22 - August 2, 2002, Koblenz, Germany
30
LAPPEENRANTA UNIVERSITY OF TECHNOLOGYTHE DEPARTMENT OF INFORMATION TECHNOLOGY
1. Infinite scope parameter space.
2. Arbitrarily high parameter resolution.
3. High computational speed.
4. Small storage.
Advances of RHT over SHT
31
LAPPEENRANTA UNIVERSITY OF TECHNOLOGYTHE DEPARTMENT OF INFORMATION TECHNOLOGY
RHT Extensions
Kälviäinen, H.,Hirvonen, P.,
Xu, L.,Oja, E.,
”Probabilistic and Non-probabilistic
Hough Transforms:Overview andComparisons,”
Image and VisionComputing,
Vol. 13, No. 4, 1995,pp. 239-251.
H. Kälviäinen, IPCV 2002, July 22 - August 2, 2002, Koblenz, Germany
32
LAPPEENRANTA UNIVERSITY OF TECHNOLOGYTHE DEPARTMENT OF INFORMATION TECHNOLOGY
Feature extraction using Hough Transform
End-pointdetection
Encoding
Input Image
Feature Image
Hough Transform
FEATURE EXTRACTION
Line parameters
Reconstruction
Feature File
July 17, 2000 H. Kälviäinen, IPCV 2002, July 22 - August 2, 2002, Koblenz, Germany
LAPPEENRANTA UNIVERSITY OF TECHNOLOGYTHE DEPARTMENT OF INFORMATION TECHNOLOGY
Applications of Hough Transform
• Randomized Hough Transform (RHT)• Curve detection• Motion detection• Mixed pixel classification• Image compression• Vanishing point detection• Image databases• etc.
July 17, 2000 H. Kälviäinen, IPCV 2002, July 22 - August 2, 2002, Koblenz, Germany
LAPPEENRANTA UNIVERSITY OF TECHNOLOGYTHE DEPARTMENT OF INFORMATION TECHNOLOGY
Application of Hough Transform for image databases
• Content-based matching of line-drawing images using Hough Transform
• Similarity of images in image databases
• Hough Transform as a feature extractor
• Translation-, rotation-, and scale-invariant features from the accumulator matrix
H. Kälviäinen, IPCV 2002, July 22 - August 2, 2002, Koblenz, Germany
35
LAPPEENRANTA UNIVERSITY OF TECHNOLOGYTHE DEPARTMENT OF INFORMATION TECHNOLOGY
July 17, 2000 H. Kälviäinen, IPCV 2002, July 22 - August 2, 2002, Koblenz, Germany
LAPPEENRANTA UNIVERSITY OF TECHNOLOGYTHE DEPARTMENT OF INFORMATION TECHNOLOGY
Query images
July 17, 2000 H. Kälviäinen, IPCV 2002, July 22 - August 2, 2002, Koblenz, Germany
LAPPEENRANTA UNIVERSITY OF TECHNOLOGYTHE DEPARTMENT OF INFORMATION TECHNOLOGY
Test database
H. Kälviäinen, IPCV 2002, July 22 - August 2, 2002, Koblenz, Germany
38
LAPPEENRANTA UNIVERSITY OF TECHNOLOGYTHE DEPARTMENT OF INFORMATION TECHNOLOGY
Image Processing Using Gabor Filtering
• For local and global feature extraction. • Filtering in time (spatial) space and frequency space.• For image processing and analysis two important
parameters: frequency f and orientation theta.• More information:
– Gabor lecture notes 1: (IPCV2002_Gabor1.ps) Introduction to the theory of Gabor functions.– Gabor lecture notes 2: (IPCV2002_Gabor2.ps) Image analysis using Gabor filtering: practice and applications.