200 likes | 220 Views
Elements of Biomedical Image Processing BMI 731 Winter 2005. Kun Huang Department of Biomedical Informatics Ohio State University. Introduction to imaging processing Mathematical background Convolution and Fourier transform Filtering Image enhancement Noise removal
E N D
Elements of Biomedical Image ProcessingBMI 731 Winter 2005 Kun Huang Department of Biomedical Informatics Ohio State University
Introduction to imaging processing • Mathematical background • Convolution and Fourier transform • Filtering • Image enhancement • Noise removal • Color correction and color space transform • Feature extraction • Edge, point, line (Hugh transform) • 3-D reconstruction • Radon transform
Image Processing : what should be done? • Image restoration and enhancement • Feature extraction • Pattern recognition
1x2+8x9+15x4+7x7+14x5 +16x3+13x6+20x1+22x8 =575 • Mathematical Background • Convolution • 2-D convolution
Mathematical Background • Fourier transform (FT) • Mathematics • 2-D FT
Mathematical Background • Fourier transform (FT) • Fast FT (FFT)
Mathematical Background • Convolution and Fourier transform (FT)
Mathematical Background • Filtering • High-pass filter, low-pass filter, band pass filter • Gradient filters
Mathematical Background • Filtering • Wiener filter and deblurring
43 • Image Enhancement • Denoise • Averaging • Median filter
Image Enhancement • Denoise/restoration From Gonzalez, Woods, and Eddins
Image Enhancement • Color and intensity adjustment • Histogram equalization
RGB -> HSV, HSL, YCbCr, … R = 64 G = 31 B = 62 R = 125 G = 80 B = 147 H = 214 S = 132 V = 64 H = 199 S = 117 V = 147 • Image Enhancement • Color space transform
Feature Extraction • Region detection – morphology manipulation • Dilate and Erode • Open • Erode dilate • Small objects are removed • Close • Dilate Erode • Holes are closed • Skeleton and perimeter
Feature Extraction • Edge detection • Gradients • Canny edge detector • Gaussian smoothing • Gradients • Two thresholds • Thinning
x • Feature Extraction • Point detection • Harris detector
y q y q • Feature Extraction • Radon transform • Straight line detection • Hugh transform
From Gonzalez, Woods, and Eddins • Feature Extraction • Straight line detection • Hugh transform
2-D/3-D reconstruction • Radon/inverse radon transforms and backprojection
Reference • Digital Image Processing using Matlab By R.C.Gonzalez, R.E.Woods, and S.L.Eddins Published by Printice-Hall, 2004