Image Enhancement.

Slides:



Advertisements
Similar presentations
Types of Image Enhancements in spatial domain
Advertisements

Digital Image Processing
Image Processing Lecture 4
Chapter 3 Image Enhancement in the Spatial Domain.
Chapter - 2 IMAGE ENHANCEMENT
Digital Image Processing
ECE 472/572 - Digital Image Processing
Image Enhancement in the Spatial Domain
Image (and Video) Coding and Processing Lecture 5: Point Operations Wade Trappe.
Digital Image Processing
Digital Image Processing In The Name Of God Digital Image Processing Lecture3: Image enhancement M. Ghelich Oghli By: M. Ghelich Oghli
Digital Image Processing
Chapter 3: Image Enhancement in the Spatial Domain
DREAM PLAN IDEA IMPLEMENTATION Introduction to Image Processing Dr. Kourosh Kiani
Image Enhancement To process an image so that the result is more suitable than the original image for a specific application. Spatial domain methods and.
Image Enhancement in the Spatial Domain (chapter 3) Math 5467, Spring 2008 Most slides stolen from Gonzalez & Woods, Steve Seitz and Alexei Efros.
Digital Image Processing
Digital Image Processing
Chapter 3 Image Enhancement in the Spatial Domain.
ECE 472/572 - Digital Image Processing Lecture 4 - Image Enhancement - Spatial Filter 09/06/11.
University of Ioannina - Department of Computer Science Intensity Transformations (Point Processing) Christophoros Nikou Digital Image.
CS654: Digital Image Analysis Lecture 17: Image Enhancement.
Chapter 3 Image Enhancement in the Spatial Domain.
Digital Image Processing Contrast Enhancement: Part I
CS6825: Point Processing Contents – not complete What is point processing? What is point processing? Altering/TRANSFORMING the image at a pixel only.
Intensity Transformations or Translation in Spatial Domain.
Intensity Transformations (Histogram Processing)
CIS 601 – 04 Image ENHANCEMENT in the SPATIAL DOMAIN Longin Jan Latecki Based on Slides by Dr. Rolf Lakaemper.
Digital Image Processing EEE415 Lecture 3
Visual Computing Computer Vision 2 INFO410 & INFO350 S2 2015
Digital Image Processing CSC331 Image Enhancement 1.
Image Enhancement in Spatial Domain Presented by : - Mr. Trushar Shah. ME/MC Department, U.V.Patel College of Engineering, Kherva.
Lecture Reading  3.1 Background  3.2 Some Basic Gray Level Transformations Some Basic Gray Level Transformations  Image Negatives  Log.
Digital Image Processing Image Enhancement in Spatial Domain
Digital Image Processing, 2nd ed. © 2002 R. C. Gonzalez & R. E. Woods Chapter 3 Image Enhancement in the Spatial Domain Chapter.
Image Enhancement in the Spatial Domain.
Lecture Six Figures from Gonzalez and Woods, Digital Image Processing, Second Edition, Copyright 2002.
Image enhancement algorithms & techniques Point-wise operations
IMAGE ENHANCEMENT TECHNIQUES
IMAGE PROCESSING INTENSITY TRANSFORMATION AND SPATIAL FILTERING
Digital Image Processing
CIS 601 – 03 Image ENHANCEMENT SPATIAL DOMAIN Longin Jan Latecki
Digital Image Processing
Histogram Histogram is a graph that shows frequency of anything. Histograms usually have bars that represent frequency of occuring of data. Histogram has.
Fundamentals of Image Processing A Seminar on By Alok K. Watve
Image Enhancement in the Spatial Domain
Lecture Five Figures from Gonzalez and Woods, Digital Image Processing, Second edition, Prentice-Hall,2002.
Image Enhancement in the
Lecture 3 (2.5.07) Image Enhancement in Spatial Domain
CSC 381/481 Quarter: Fall 03/04 Daniela Stan Raicu
Image Enhancement Gray level transformation Linear transformation
Digital Image Processing
Histogram Probability distribution of the different grays in an image.
Digital Image Processing
Digital Image Processing
Image Processing Ch3: Intensity Transformation and spatial filters
Lecture Four Chapter Three
Chapter 3 Image Enhancement in the Spatial Domain
Grey Level Enhancement
Image Enhancement To process an image so that the result is more suitable than the original image for a specific application. Spatial domain methods and.
CIS 4350 Image ENHANCEMENT SPATIAL DOMAIN
Intensity Transformations and Spatial Filtering
IT523 Digital Image Processing
Intensity Transform Contrast Stretching Y ← u0+γ*(Y-u)/s
IT523 Digital Image Processing
The spatial domain processes discussed in this chapter are denoted by the expression
Image Enhancement in the Spatial Domain
DIGITAL IMAGE PROCESSING Elective 3 (5th Sem.)
Presentation transcript:

Image Enhancement

References Gonzalez and Woods, “ Digital Image Processing,” 2nd Edition, Prentice Hall, 2002. Jahne, “Digital Image Processing,” 5th Edition, Springer 2002. Jain, “Fundamentals of Digital Image Processing,” Prentice Hall 1989 9/19/2018

Overview Human perception (focus of this discussion) Machine perception (ocr) Application specific Heuristic based: result better than the original image – subjective assessment Spatial vs frequency domain 9/19/2018

Spatial Domain Based on the collection of pixels in the image Enhancement techniques will yield Noise reduction Neighborhood smoothing Highlighting of desired feature 9/19/2018

Spatial Domain – Math Framework From [1] Typically, spatial domain based enhancement involves: g(x,y) = T [f(x,y)], where f = input image; g = output image; T = operator defined on f based on a neighborhood of x,y. If the neighborhood is 1x1 pixel, then the output intensity is dependent on the current intensity value of the pixel, and can be represented as s = T( r ) where r and s are gray level values of f(x,y) and g(x,y) at location x,y. In such situations T is a gray-level transformation function. 9/19/2018

Examples of Gray level Transformation Functions (Point Processing) From [1] Contrast Stretching Best if input is 0 outside a range of values. Thresholding: Result is a binary Image 9/19/2018

Larger Neighborhoods Objective – determine g(x,y) based on input intensity (gray level) values f(x,y) in the neighborhood of x,y. Mask processing or filtering Each of the elements in the neighborhood has an associated weight g(x,y) depends on f(a,b)|a,b є N(x,y) 9/19/2018

Basic Gray Level Transformations Light From [1] Dark Dark Light 9/19/2018

Image Negatives From [1] In this example, using the image negative, it is easier to analyze the breast tissue. 9/19/2018

Log transformations s = c log (1 + r) r >= 0; hence 1 + r > 0; log 0 = ? Log transformations are useful, when there is a large dynamic range for the input variable ( r ). From [1] 9/19/2018 Range: 0 to 1.5*106 Range: 0 to 6.2

Power law transformation From [1] Stretch higher (lighter) gray levels Many display devices (e.g. CRT) respond like the power law, i.e intensity – voltage relationship is power law based gamma 1.8 to 2.5. The display will tend to produce images darker than intended. So the display is distorted. Gamma correction is used to correct for this distortion. Stretch lower (ldarker) gray levels 9/19/2018

Display distortion correction Gamma correction can also fix the distortions in color. More important with the internet. Many viewers, variety of monitors. Gamma of view station is not known. Preprocess using an average gamma. From [1] 9/19/2018

Power law – contrast manipulation (c) better than (b). (d) Background is better than (c) but washed out effect. From [1] 9/19/2018

Piece-wise Linear Transformation Contrast Stretching From [1] Mean gray level value of the image 9/19/2018

Gray level Slicing From [1] 9/19/2018

Bit plane Slicing MSb (bit 7) From [1] 9/19/2018 LSb (bit 0)

Histograms Histogram - frequency of occurrence of a gray level value Normalizing histograms with rest to the total number of pixels converts these into probability density like function Histogram processing yields robust image processing results Histograms are NOT unique 9/19/2018

Histograms for 4 images From [1] For high contrast, it is best to have a larger range of gray level values. If we could transform an image with a resulting change in histogram, then that may yield more contrast. We need to study the rules for transforming histograms, and study the resulting impact on images. 9/19/2018 From [1]

Transformation From [1] s s1 r r s1 s Applying transformations to histograms, can use results from probability. Consider the transformation S = T( r ); 0 <= r <=1 which has the following properties: T( r ) is single-valued and monotonically increasing in [0,1] 0 <= T( r ) <= 1 for r in [0,1] Single valued condition ensures that an inverse transformation exists, and the monotonic condition ensures the increasing order of the pixels from black to white. (a) and (b) do not ensure that the inverse transform is single valued. s1 r r s1 s 9/19/2018

Histogram Equalization 9/19/2018 From [1]

Transformation Functions From [1] 9/19/2018

Mapping for Histogram Specification From [1] 9/19/2018

Example of Histogram Specification From [1] 9/19/2018

Continued From [1] 9/19/2018

Localized Histogram Equalization From [1] 9/19/2018

From [1] 9/19/2018

Histogram Stats From [1] 9/19/2018