340 likes | 549 Views
A Seminar on Fundamentals of Image Processing By Alok K. Watve. Applications of image processing. Gamma ray imaging X-ray imaging Multimedia systems Satellite imagery Flaw detection and quality control And many more……. Fundamental Steps in digital image processing. Image acquisition
E N D
A Seminar on Fundamentals of Image Processing By Alok K. Watve
Applications of image processing • Gamma ray imaging • X-ray imaging • Multimedia systems • Satellite imagery • Flaw detection and quality control • And many more…….
Fundamental Steps in digital image processing • Image acquisition • Image enhancement(gray or color images) • Wavelet and multi-resolution processing • Compression • Morphological processing • Segmentation • Representation & description • Object recognition Low level processing Medium level processing High level processing
Image enhancement in spatial domain • Binary images • Only two colors • Gray images • A range of colors(not more than 256) from black to white • Color images • Contain several colors(as many as 224)
Definitions • Image : A 2D function represented by • I = f(x,y) • where I = intensity of the point(x,y) • Foreground : Objects of interest in an image • Background : Everything that’s not in foreground
Definitions • Histogram : A graph of frequency(of an intensity) versus intensity. Frequency is expressed as count of pixels • freq(I) = # pixels with intensity I • Spatial resolution : Smallest discernible detail in the image. Depends on the sampling. • Gray-level resolution : smallest discernible change in the gray level change.
Basic gray level transformations Image negatives s = (L – 1) – r Where, s = output intensity* r = input intensity* (L – 1) = Maximum intensity* *These notations will be used throughout the seminar
Basic gray level transformations Original image Negative image All images: courtesy : www.imageprocessingplace.com
Basic gray level transformations Power law transformation s = c. rγ Here, c is a constant
Basic gray level transformations Original image Transformed image (c = 1, γ = 0.3)
Basic gray level transformations Contrast stretching : increases dynamic range L-1 S2 Output intensity S1 0, 0 L1 L2 L-1 Input intensity
Basic gray level transformations Original image Image obtained by contrast stretching
Basic gray level transformations Bit plane slicing
Basic gray level transformations Histogram equalization Image of mars’ moon histogram
Basic gray level transformations Histogram equalization transformation can be expressed as a monotonically increasing function with domain and range = [0, 1]** Assuming the intensities are normalized in the range [0,1]
Basic gray level transformations A low contrast image and its histogram
Basic gray level transformations Result of histogram equalization
Filtering in spatial domain • Concept of frequency • Modeling filters using convolution in spatial domain • Implementing filters using masks
g(x, y) = ΣsΣt w(s, t).f(x+s,y+t) ΣsΣt w(s, t)
Filtering in spatial domain w(-1,-1) w(-1, 0) w(-1, 1) w(0,-1) w(0, 0) w(0, 1) f(x-1,y-1) f(x-1, y) f(x-1, y+1) w(1,-1) w(1, 0) w(1, 1) f(x,y-1) f(x, y) f(x, y+1) f(x+1,y-1) f(x+1, y) f(x+1, y+1)
Low pass filter Averaging filter Weighted average filter
Median filter A noisy image Filtered image
High pass filter • Computing gradients in spatial domain • Laplacian filter • Other masks (operators) • Roberts • Sobel
High pass filter • Designing high pass filters • Method 1 • g(x, y) = f(x, y) + fhp(x, y) • Method 2 • g(x, y) = f(x, y) – flp(x, y)
Laplacian operators 0 -1 0 -1 -1 -1 -1 5 -1 -1 9 -1 0 -1 0 -1 -1 -1
Sobel Operators -1 -2 -1 -1 0 1 0 0 0 -2 0 2 1 2 1 -1 0 1
Roberts cross gradient operators - 1 0 0 -1 0 1 1 0
References • Digital image processing, second edition - R. C. Gonzalez, R. E. Woods • Fundamental of digital image processing – A. K. Jain • www.imageprocessingplace.com