Basic Image processing
techniques
H Anandkumar Singh
Student : @ Veltech Dr RR & Dr SR Technical University

Image Processing
• A Digital Image Processing System--- Java open CV or Matlab C program or MATLAB
• Image Representation and Formats
1. Sensing, Sampling, Quantization
2. Gray level and Color Images
3. Raw, RGB, Tiff, BMP, JPG, GIF, (JP2)
• Image Transform and Filtering
• Histogram, Enhancement and Restoration
• Segmentation, Edge Detection, Thinning
• Image Data Compression
• Image Pattern Analysis (Recognition and Interpretation)
[1] R.C. Gonzalez, R.E. Woods, S.L. Eddins, Digital Image Processing Using MATLAB, Pearson Prentice
Hall, 2004
[2] R.C. Gonzalez and R.E. Woods, Digital Image Processing, Prentice-Hall, 2002+

Image Processing System

Digital Image Analysis System
• A 2D image is nothing but a mapping from a region to a matrix
• A Digital Image Processing System consists of
1. Acquisition – scanners, digital camera, ultrasound,
X-ray, MRI
2. Storage – HD (120GB), CD (700MB), DVD (4.7GB),
Flash memory (512MB~4GB), …
3. Processing Unit – PC, Workstation, PC-cluster
4. Communication – telephone lines, cable, wireless, …
5. Display – LCD monitor, laser printer, laser-jet printer

Gray Level and Color Images

Simple Code to read Image and display in
Matlab
• %% Reading Image of 2D
• I = imread('circuit.tif'); %Reading the Image and stores in Variable I
• figure, subplot(121)% Define a location on the figure to display
• imshow(I,[]);% displaying the Read Image on the define location
• %% Reading Colour Image
• I = imread('peppers.png');
• subplot(122)
• imshow(I,[]);

Pixels in a Gray Level Image

Pixels Value(0~255) Gray Level Image
15 16 15 14 15 16 15
15 27 56 66 73 81 83
15 24 57 63 66 72 81
15 22 47 59 66 66 81
14 20 43 56 64 63 75
15 16 42 49 56 62 68
14 15 34 41 51 59 58

A Gray Level Image is a Matrix
f(0,0) f(0,1) f(0,2) …. …. f(0,n-1)
f(1,0) f(1,1) f(1,2) …. …. f(1,n-1)
. . .
. . .
. . .
f(m-1,0) f(m-1,1) f(m-1,2) … …. f(m-1,n-1)
An image of m rows, n columns, f(i,j) is in [0,255]

Red, Green, Blue, Color Images

Sensing, Sampling, Quantization
• A 2D digital image is formed by a sensor which maps a region to a
matrix
• Digitization of the spatial coordinates (x,y) in an image function f(x,y)
is called Sampling
• Digitization of the amplitude of an image function f(x,y) is called
Quantization

Sensing, Sampling, Quantization

Image File Formats
• The most common image file formats, the
most important for cameras, printing,
scanning, and internet use, are JPG, TIF,
PNG, and GIF.
• DICOM image for Medical Image
processing
• Common Image FileFormat:
BIN, RAW
PPM,PBM,PGM
BMP
JPEG
TIFF
GIF
RAS
SGI
PNG
PICT, FPX
EPS
VIP
• Satellite Image file format:
• CCRS format
• EOSAT Fast Format
• Fast-L7A Format
• HDF Format

Filtering
• Median Filter,
• Low Pass filter, High Pass filter, Band Pass filter,
• Gabor Filter, Log Gabor filter,
• Normalized 2-D cross-correlation filter,
• 2-D adaptive noise-removal filtering
• 2-D FIR filter using frequency sampling
• Local range filter of image
• Local standard deviation of image
• Local entropy of grayscale image
Consists of three types of filters:
• Mean filters
• Median filters
• Enhancement filters

Laplacian Filter
 Are rotationally invariant , that is they enhance the details in
all directions equally
 Example convolution masks of Laplacian-type filters are:

Image Transformation
•Feature Extraction – find all ellipses in an image
•Bandwidth Reduction – eliminate the low
contrast “coefficients”
•Data Reduction – eliminate insignificant
coefficients of Discrete Cosine Transform (DCT),
Wavelet Transform (WT)
•Smooth filtering can get rid of noisy signals

Matlab Image Transformation
•Discrete Cosine Transform
•Fast Fourier Transform
•Wavelet Transform
•Radon Transform
•Hough Transform
•Distance Transform

Matlab Variable
• 2-D Convolution Compute 2-D discrete convolution of two input matrices
• 2-D FFT Compute two-dimensional fast Fourier transform of input
• 2-D IFFT 2-D Inverse fast Fourier transform of input
• 2-D DCT Compute 2-D discrete cosine transform (DCT)
• 2-D IDCT Compute 2-D inverse discrete cosine transform (IDCT)
• 2-D FIR Filter Perform 2-D FIR filtering on input matrix
• Contrast Adjustment Adjust image contrast by linearly scaling pixel values
• Deinterlacing Remove motion artifacts by deinterlacing input video signal
• Edge Detection Find edges of objects in images using Sobel, Prewitt, Roberts, or Canny method
• Histogram Equalization Enhance contrast of images using histogram equalization
• Median Filter Perform 2-D median filtering
• Hough Transform Find lines in images
• Hough Lines Find Cartesian coordinates of lines described by rho and theta pairs
• Gaussian Pyramid Perform Gaussian pyramid decomposition

Enhancement and Restoration
•The goal of enhancement is to accentuate
certain features for subsequent analysis or
image display. The enhancement process is
usually done interactively
•The restoration is a process that attempts to
reconstruct or recover an image that has been
degraded by using some unknown phenomenon

Image Enhancement
• Grayscale colormap for contrast enhancement
• Matlab Library:
• vision.ContrastAdjuster Adjust image contrast by linear scaling
• vision.Convolver Compute 2-D discrete convolution of two input matrices
• vision.FFT Two-dimensional discrete Fourier transform
• vision.IFFT Two–dimensional inverse discrete Fourier transform
• vision.DCT Compute 2-D discrete cosine transform
• vision.IDCT Compute 2-D inverse discrete cosine transform
• vision.Deinterlacer Remove motion artifacts by deinterlacing input video signal
• vision.EdgeDetector Find object edge
• vision.HistogramEqualizer Enhance contrast of images using histogram equalization
• vision.HoughTransform Find lines in images via Hough transform
• vision.HoughLines Find Cartesian coordinates of lines that are described by rho and theta pairs
• vision.ImageFilter Perform 2-D FIR filtering of input matrix
• vision.MedianFilter 2D median filtering
• vision.Pyramid Perform Gaussian pyramid decomposition

Feature Detection and Extraction
•Interest point detection,
•extracting feature descriptors,
•Point feature matching

Functions
• detectBRISKFeatures -Detect BRISK features and return BRISKPoints object
• detectFASTFeatures- Detect corners using FAST algorithm and return cornerPoints
object
• detectHarrisFeatures- Detect corners using Harris–Stephens algorithm and return
cornerPoints object
• detectMinEigenFeatures- Detect corners using minimum eigenvalue algorithm and
return cornerPoints object
• detectMSERFeatures- Detect MSER features and return MSERRegions object
• detectSURFFeatures- Detect SURF features and return SURFPoints object
• extractFeatures- Extract interest point descriptors
• extractHOGFeatures- Extract histogram of oriented gradients (HOG) features
• matchFeatures- Find matching features
• showMatchedFeatures- Display corresponding feature points

Image Restoration--- Wavelet Decomposition
• Restoration by Fusion from Fuzzy Images
• example
• Restoration by fussing two different images
• Example
• Image Restoration --Deblur image using Wiener filter
• Example
• Deblur image using Lucy-Richardson method
• example

Denoising
• Functions
• cmddenoise Interval-dependent denoising
• ddencmp Default values for denoising or compression
• thselect Threshold selection for de-noising
• wbmpen Penalized threshold for wavelet 1-D or 2-D de-noising
• wdcbm Thresholds for wavelet 1-D using Birgé-Massart strategy
• wdcbm2 Thresholds for wavelet 2-D using Birgé-Massart strategy
• wden Automatic 1-D de-noising
• wdencmp De-noising or compression
• wmulden Wavelet multivariate de-noising
• wnoise Noisy wavelet test data
• wnoisest Estimate noise of 1-D wavelet coefficients
• wpbmpen Penalized threshold for wavelet packet de-noising
• wpdencmp De-noising or compression using wavelet packets
• wpthcoef Wavelet packet coefficients thresholding
• wthcoef 1-D wavelet coefficient thresholding
• wthcoef2 Wavelet coefficient thresholding 2-D
• wthresh Soft or hard thresholding
• wthrmngr Threshold settings manager
• wvarchg Find variance change points
• measerr Approximation quality metrics
• wavemenu Wavelet Toolbox GUI tools

Image Compression
• Functions
• imgCompress Compress HDU from one file into another
• isCompressedImg Determine if current image is compressed
• setCompressionType Set image compression type
• setHCompScale Set scale parameter for HCOMPRESS
algorithm
• setHCompSmooth Set smoothing for images compressed with
HCOMPRESS
• setTileDim Set tile dimensions

Image Compression
• Two-Dimensional True Compression
• example
• Compression by Global Thresholding and Huffman Encoding
• example
• Uncompression
• Compression by Progressive Methods
• Handling Truecolor Images
• More about True Compression of Images

Segmentation and Edge Detection
• Segmentation is basically a process of pixel classification: the picture
is segmented into subsets by assigning the individual pixels into
classes
• Edge Detection is to find the pixels whose gray values or colors being
abruptly changed

Image Segmentation
• Functions in Matlab
• activecontour Segment image into foreground and background using active
contour
• imsegfmm Binary image segmentation using Fast Marching Method
• gradientweight Calculate weights for image pixels based on image gradient
• graydiffweight Calculate weights for image pixels based on grayscale intensity
difference
• graythresh Global image threshold using Otsu's method
• multithresh Multilevel image thresholds using Otsu's method

Edge Detection
• Edge finding Syntax;
• BW = edge(I)
• gpuarrayBW = edge(gpuarrayI)
• BW = edge(I,'sobel')
• BW = edge(I,'sobel',thresh)
• BW = edge(I,'sobel',thresh,direction)
• BW = edge(I,'sobel',...,options)
• [BW,thresh] = edge(I,'sobel',...)
• BW = edge(I,'prewitt')
• BW = edge(I,'prewitt',thresh)
• BW = edge(I,'prewitt',thresh,direction)
• [BW,thresh] = edge(I,'prewitt',...)
• BW = edge(I,'roberts')
• BW = edge(I,'roberts',thresh)
• BW = edge(I,'roberts',...,options)
• [BW,thresh] = edge(I,'roberts',...)
• BW = edge(I,'log')
• BW = edge(I,'log',thresh)
• BW = edge(I,'log',thresh,sigma)
• [BW,thresh] = edge(I,'log',...)
• BW = edge(I,'zerocross',thresh,h)
• [BW,thresh] = edge(I,'zerocross',...)
• BW = edge(I,'canny')
• BW = edge(I,'canny',thresh)
• BW = edge(I,'canny',thresh,sigma)
• [BW,thresh] = edge(I,'canny',...)

Edge Detection
• Hough Transform
• Corner Detection
• Boundary Tracing in Images
• Quadtree Decomposition

GUI
• Create a Simple GUIDE GUI

• List Pop-Up Menu Items
• The pop-up menu provides a choice of three data sets: peaks, membrane, and sinc. These data sets correspond to MATLAB functions of the
same name. This topic shows you how to list those data sets as choices in the pop-menu.
• In the layout area, click the pop-up menu.
• In the Property Inspector, click the button next to String. The String dialog box displays.

Replace the existing text with the names of the three data sets: Peaks,
Membrane, and Sinc. Press Enter to move to the next line.

When you finish editing the items, click OK.

Modify the Static Text
In this GUI, the static text serves as a label for the pop-up menu. This topic shows you how to change the static text to read Select Data.
In the layout area, click the static text.
In the Property Inspector, click the button next to String. In the String dialog box that displays, replace the existing text with the phrase Select
Data.

• Save the GUI Layout
• When you save a GUI, GUIDE creates two files, a FIG-file and a code file. The FIG-file, with extension .fig, is a binary file that
contains a description of the layout. The code file, with extension .m, contains MATLAB functions that control the GUI behavior.
• Save and activate your GUI by selecting Tools > Run.
• GUIDE displays a dialog box displaying: "Activating will save changes to your figure file and MATLAB code. Do you wish to continue?
• Click Yes.
• GUIDE opens a Save As dialog box in your current folder and prompts you for a FIG-file name.
• Browse to any folder for which you have write privileges, and then enter the file name simple_gui for the FIG-file. GUIDE saves
both the FIG-file and the code file using this name.
• If the folder in which you save the GUI is not on the MATLAB path, GUIDE opens a dialog box, giving you the option of changing the
current folder to the folder containing the GUI files, or adding that folder to the top or bottom of the MATLAB path.
• GUIDE saves the files simple_gui.fig and simple_gui.m, and then activates the GUI. It also opens the GUI code file in your default
editor.
• The GUI opens in a new window. Notice that the GUI lacks the standard menu bar and toolbar that MATLAB figure windows
display. You can add your own menus and toolbar buttons with GUIDE, but by default a GUIDE GUI includes none of these
components.
• When you run simple_gui, you can select a data set in the pop-up menu and click the push buttons, but nothing happens. This is
because the code file contains no statements to service the pop-up menu and the buttons.

Code the Simple GUIDE GUI Behavior
• % Create the data to plot.
• handles.peaks=peaks(35);
• handles.membrane=membrane;
• [x,y] = meshgrid(-8:.5:8);
• r = sqrt(x.^2+y.^2) + eps;
• sinc = sin(r)./r;
• handles.sinc = sinc;
• % Set the current data value.
• handles.current_data = handles.peaks;
• surf(handles.current_data)

• % Determine the selected data set.
• str = get(hObject, 'String');
• val = get(hObject,'Value');
• % Set current data to the selected data set.
• switch str{val};
• case 'Peaks' % User selects peaks.
• handles.current_data = handles.peaks;
• case 'Membrane' % User selects membrane.
• handles.current_data = handles.membrane;
• case 'Sinc' % User selects sinc.
• handles.current_data = handles.sinc;
• end
• % Save the handles structure.
• guidata(hObject,handles)

• Add the following code to the callback immediately after the comment that
begins % handles...
• % Display surf plot of the currently selected data.
• surf(handles.current_data);
• Add this code to the Mesh push button callback, pushbutton2_Callback:
• % Display mesh plot of the currently selected data.
• mesh(handles.current_data);
• Add this code to the Contour push button callback, pushbutton3_Callback:
• % Display contour plot of the currently selected data.
• contour(handles.current_data);
• Save your code by selecting File > Save.

• handles.Dataset = 'Databasetrain';
•
• k = dir(fullfile(handles.Dataset,'*.bmp')); % Read All file in(Dataset =
'Databasepalmprint';)
• k = {k(~[k.isdir]).name}; % Read Filename extension
• ------------------------
• [FileName,PathName] = uigetfile('*.m','Select the MATLAB code file');

The session Is ended
• Any queries please !!!!!!!!!!!!!!!!!!!!