Image Processing

1
Image Processing
2
What is Image Processing ?

• Image Processing is any form of signal
  processing for which our input is an image, such
  as photographs or frames of video and our output
  can be either an image or a set of
  characteristics or parameters related to the
  image.

3

• Image Processing generally refers to processing
  of two dimensional picture and by two dimensional
  picture we implies a digital image.
• A digital image is an array of real or complex
  numbers represented by a finite number of bits.
• But now in these days optical and analog image
  processing is also possible.

4
Applications

•   Face detection
• Feature detection
• Non-photorealistic rendering
• Medical image processing
• Microscope image processing
• Morphological image processing
• Remote sensing
• Automated Sieving Procedures
• Finger print recognization

5
How image processing is done ?

• Image processing can be done using various
  softwares and languages such as-
• Software
• Matlab
• Adobe photoshop
• Irfan view
• Language
• VHDL
• C/C

6
Image processing using MATLAB
7
MATLAB

• The name MATLAB stands for matrix laboratory .
  It is a high-performance language for technical
  computing. It is an interactive system whose
  basic data element is an array which does not
  require any dimensioning. This allows us to solve
  many technical computing problems, especially
  those with matrix and vector formulations, in a
  fraction of the time.

8

• MATLAB features a family of add-on
  application-specific solutions called toolboxes.
  These toolboxes are comprehensive collections of
  matlab functions (M-files) that extend the matlab
  environment to solve particular classes of
  problems. Areas in which toolboxes are available
  include image processing, signal processing,
  control systems, neural networks, fuzzy logic,
  wavelets, simulation, and many others..

9
Image Processing Toolbox
10
Image processing toolbox

• The Image Processing Toolbox is a collection of
  functions that extend the capability of the
  Matlab numeric computing environment. These
  functions are written on M-files and further we
  can extend the capabilities of the Image
  Processing Toolbox by writing our own M-files.
  This toolbox supports a wide range of image
  processing operations such as..

11
Operations of image processing

• Color corrections such as brightness and contrast
  adjustments, quantization, or color translation
  to a different color space.
• Image registration, the alignment of two or more
  images.
• Image segmentation.

12

• Neighborhood and block operations.
• Linear filtering and filter design.
• Transforms.
• High dynamic range imaging by combining multiple
  images.
• Deblurring.

13
Image formats supported by Matlab

• The following image formats are supported by
  Matlab
•  
• BMP
• HDF
• JPEG
• PCX
• TIFF
• XWB

14
Types of images

• Intensity image
• This is the equivalent to a "gray scale image .
  It represents an image as a matrix where every
  element has a value corresponding to how
  bright/dark (each element represent intensities).
  
• Binary image
• This image format also stores an image as a
  matrix but can only color a pixel black or white
  (and nothing in between). It assigns a 0 for
  black and a 1 for white.

15

• Indexed image
• This is a practical way of representing color
  images. An indexed image stores an image as two
  matrices. The first matrix has the same size as
  the image and one number for each pixel. The
  second matrix is called the color map and its
  size may be different from the image. The numbers
  in the first matrix is an instruction of what
  number to use in the color map matrix.

16

• RGB image
• It represents an image with three matrices of
  sizes matching the image format. Each matrix
  corresponds to one of the colors red, green or
  blue and gives an instruction of how much of each
  of these colors a certain pixel should use. A
  pixel whose color components are (255,255,255) is
  displayed as White. And whose color components
  are (0,0,0) is displayed as Black.

17
Commands for intensity Adjustment

• Histeq( ) This command is used to improve the
  contrast in the image. This function spreads the
  intensity values over the full range of the
  image, and this process is known as histogram
  equalization.
• Syntax
• histeq(Image_name.format)

After
Before
18

• Imadjust( ) This command is used to adjust the
  contrast of the image. Imadjust increases the
  contrast of the image by saturating 1 of the
  data at both low and high intensities of Image
  and by stretching the intensity.
• Syntax
• Imadjust(image_name.format)

19
Examples of imadjust
20

• Adapthisteq( ) This command is used to perform
  contrast-limited adaptive histogram equalization
  (CLAHE). It is an alternative to function
  histeq, While histeq works on the entire image,
  adapthisteq operates on small regions in the
  image, called tiles.
• Syntax
• adapthisteq(Image_name.format)

Before
After
21
Commands used for various file operations in
image processing toolbox.

• Imread( ) This command is used to read that
  image on which the operation has to be done.
  Imread returns the image data in the array. If
  the file contains a grayscale image, then it will
  return a two-dimensional (M-by-N) array and if
  the file contains a color image, it will return a
  three-dimensional (M-by-N-by-3) array. The class
  of the returned array depends on the data type
  used by the file format.
• Syntax
• Imread(image_name.format)

22

• Imview( ) This command is used to view the image
  on the screen. And this command is always used
  with the imread and imwrite command, because, it
  views only that image which is under process.
• Syntax
• Imview(image_name.format)

23

• Result of above explained commands.

24

• Imwrite( , ) This command is used to change the
  format of read file. By using this command we can
  change the file formats from one to another.
• Syntax
• Imwrite(image_name.format2,
  image_name.format1)
• Imfinfo( ) This command is used to obtain the
  information about the image under process. It is
  used to obtain information such as size, position
  of pixels, number of rows and columns etc.
• Syntax
• Iminfo(image_name.format)

25

• Imsubtract( , ) This command is used to create a
  more uniform background, by subtracting the
  background image from the original image.
• To estimate the background image, we use
  following command
• backgroundimopen(Image_name.format,strel('dis
  k',15))
• To see the estimated background image, type
  imview(background)
• Now subtract the background image from the
  original image, type
• imsubtract(image_name.format,background)

26

• Result of the above used commands

27

• Imresize( , ) This command is used to
  resize the processed image. To enlarge an image,
  specify a magnification factor greater than 1 in
  the command. To reduce an image, specify a
  magnification factor between 0 and 1 in the
  command.
• Syntax
• imresize(Image_name.format, value)

28

• Image Rotation( , ) This command is used to
  rotate the given image. This command accepts two
  primary arguments, one is the image to be rotated
  and other one is rotation angle. We have to
  specify the rotation angle in degrees. If a
  positive value is specified then imrotate rotates
  the image counterclockwise and if a negative
  value then imrotate rotates the image clockwise.
• Syntax
• imrotate(Image_name.format, angle_in_degree)

29
Result of the above used commands
30

• Imcrop( , ) This command is used to crop the
  particular image. It accepts two primary
  arguments
• The image to be cropped.
• The coordinates of a rectangle that defines the
  crop area.
• If we call imcrop without specifying the crop
  rectangle, we can specify the crop rectangle
  interactively. In this case, the cursor changes
  to crosshairs when it is over the image. Position
  the crosshairs over a corner of the crop region
  and press and hold the left mouse button. When we
  drag the crosshairs over the image we specify the
  rectangular crop region. imcrop draws a rectangle
  around the area we are selecting. When we release
  the mouse button, imcrop creates a new image from
  the selected region.

31
 
Syntax Imcrop(image_name.format)
After
Before
If we call imcrop with specifying the crop
rectangle, imcrop operates on the image in the
current axes.Syntax Imcrop(Image_name.format
, rect)
32

• Imerode( , ) This command is used to erode the
  image. This function accepts two primary
  arguments
• The input image to be processed (grayscale,
  binary image),
• A structuring element object, returned by the
  strel function, or a binary matrix defining the
  neighborhood of a structuring element.
• Syntax
• Imerode(binary_image.format , strel)

Before
After
33

• Im2bw This command is used to convert the given
  image in to binary image. This command is mainly
  followed by one another command i.e.
  levelgraythresh(image_name.format).
• Syntax
• Levelgraythresh(image_name.format)
• Bwim2bw(image_name.format, level)

Before
After
34

• Duller This command is used to fade the image,
  so that, other image on which it is overlapped
  can be easily viewed.
• Syntax
• Duller image_name.format
• This is the amount by which the fading is
  done.
• Combine This command is used to combine or
  overlap the two images. This is the command on
  which over all project stands.
• Syntax
• Combine image1 image2.

0.5
35
Image Conversions

• We can convert images in any of formats/types
  described above using the following commands.
• Intensity format to Indexed format. gray2ind(
  )

Gray
Indexed
36

• indexed format to intensity format.
  ind2gray()

Indexed
Gray
37

• RGB format to intensity format. rgb2gray( )

RGB
Gray
38

• RGB format to indexed format. rgb2ind( )

RGB
Indexed
39
IMAGE ANALYSIS

• Image analysis return information about the
  structure of an image. This section describes
  toolbox functions that we can use for these image
  analysis techniques which includes
• Edge Detection
• Boundary Tracing
• Quadtree Decomposition

40
Edge Detection

• We can use the edge function to detect edges,
  which are those places in an image that
  correspond to object boundaries. To find edges,
  this function looks for places in the image where
  the intensity changes rapidly. Edge takes an
  intensity image I as its input, and returns a
  binary image BW of the same size as I, with 1's
  where the function finds edges in I and 0's
  elsewhere.

41

• Canny Method
• The most powerful edge-detection method that
  edge provides is the Canny method. The Canny
  method differs from the other edge-detection
  methods in that it uses two different thresholds
  (to detect strong and weak edges), and includes
  the weak edges in the output only if they are
  connected to strong edges. This method is
  therefore less likely than the others to be
  fooled by noise, and more likely to detect true
  weak edges. 
• Syntax
• Edge (image_name.fomat,'canny')

Before
After
42
Boundary Tracing

• The toolbox includes two functions you can use
  to find the boundaries of objects in a binary
  image
• Bwtraceboundary
• Bwboundaries

43

• Bwtraceboundary
• The bwtraceboundary function returns the row
  and column coordinates of all the pixels on the
  border of an object in an image. We must specify
  the location of a border pixel on the object as
  the starting point for the trace.
• Syntax
• boundary bwtraceboundary(binary_image,row,
  col,'N')

44

• Bwboundaries
• The bwboundaries function returns the row and
  column coordinates of border pixels of all the
  objects in an image.
• Syntax
• binary_image _filled imfill(binary_image,'holes'
  )
• boundaries bwboundaries(binary_image _filled)

45
Example of above explained commands
46
Quadtree Decomposition

• Quadtree decomposition is an analysis technique
  that involves subdividing an image into blocks
  that are more homogeneous than the image itself.
  This technique reveals information about the
  structure of the image.
• Syntax
• qtdecomp(Image_name.format,threshold)

Before
After
47
Noise Removal

• Digital images are prone to a variety of types
  of noise. There are several ways that noise can
  be introduced into an image, depending on how the
  image is created. For example
• If the image is scanned from a photograph made
  on film, the film grain is a source of noise.
  Noise can also be the result of damage to the
  film, or be introduced by the scanner itself.
• If the image is acquired directly in a digital
  format, the mechanism for gathering the data
  (such as a CCD detector) can introduce
  noise.Electronic transmission of image data can
  introduce noise.
• The Image Processing toolbox provides a number
  of different ways to remove or reduce noise in an
  image. Different methods are better for different
  kinds of noise. The method which is used mostly
  to remove noise is Filtering.

48
Filtering

• Filtering is a technique for modifying or
  enhancing an image. It is a neighborhood
  operation, in which the value of any given pixel
  in the output image is determined by applying
  some algorithm to the values of the pixels in the
  neighborhood of the corresponding input pixel. A
  pixel's neighborhood is some set of pixels,
  defined by their locations relative to that
  pixel. Filtering is further categories as
  following
• Linear Filtering
• Median Filtering
• Adaptive Filtering

49
Image Histogram

• An image histogram is a chart that shows the
  distribution of intensities in an indexed or
  intensity image. The image histogram function
  imhist creates this plot by making n equally
  spaced bins, each representing a range of data
  values. It then calculates the number of pixels
  within each range.
• Syntax
• Imhist(Image_name.format)

50
Histogram of a given image
51
Project fingerprint Recognition
52
Project Fingerprint Recognition

• Fingerprint Recognition or Fingerprint
  Authentication refers to the automated method of
  verifying a match between two human fingerprints.
  Fingerprints are one of many forms of biometrics
  used to identify an individual and verify their
  identity.
• Fingerprint Recognition touches on two major
  classes of algorithms, one is based upon PATTERNS
  and other one is based upon MINUTIA.

53
Patterns of Fingerprints

• The basic patterns of fingerprint ridges
  includes -
• Arch pattern An arch is a pattern where the
  ridges enter from one side of the finger, rise in
  the center forming an arc, and then exit the
  other side of the finger as shown in following
  figure..

54

• Loop Pattern It is a pattern where the ridges
  enter from one side of a finger, form a curve,
  and tend to exit from the same side they enter as
  shown in the following figure..

55

• Whirl Pattern It is completely different from
  others. In this ridges form circularly around a
  central point on the finger as shown in the
  following figure..

56
Minutia Features

• The major Minutia features of fingerprint ridges
  includes
• Ridge Ending The ridge ending is the point at
  which a ridge terminates and this could be better
  explained with the help of image, which is shown
  below..

57

• Bifurcations These are the points at which a
  single ridge splits into two ridges as shown in
  the following image..

58

• Short ridges (or dots) These are ridges which
  are significantly shorter than the average ridge
  length on the fingerprint as shown in the
  following image..

Short Ridge (or Dot)
59
Methods of Fingerprint Recognition

• There are various methods of fingerprint
  recognization using MATLAB which includes-
• Overlapping or Combining.
• Comparing slopes of ridges.
• Other methods.

60
Overlapping or Combining Method
61

• Overlapping or Combining Method
• It is the one of the oldest technique of
  fingerprint matching. In this the two images of
  known fingerprints and unknown fingerprints are
  overlapped and checked for similarities. Also for
  recognization the position of the pixels of
  combined image is matched with the other images
  for the better result.

62

• Basically for overlapping the images, firstly
  they are converted into the binary images and
  then they are overlapped. Also for improving the
  results some other operation are also done like
  normalization, thinning etc.. according to the
  requirements.

63
Program for overlapping
64

• Program
• iimread('fp1.bmp')
• imview(i)
• levelgraythresh(i)
• bwim2bw(i,level)
• imview(bw)
• imview(bw)
• imshow(bw,1 0 0 0 0 1)

65

• Result of the above used commands

66

• i2imread('fp2.bmp')
• imview(i2)
• levelgraythresh(i)
• bw2im2bw(i2,level)
• imview(bw2)
• imview(bw2)
• imshow(bw2,1 0 0 0 0 1)

67
Result of the above used commands
68

• duller bw2 0.5
• imview(duller)
• combinebwduller
• imview(combine)

69
Result

• From the above program, we come to know that
  whether two finger prints are similar or not.
• As it is clear from the above program that two
  fingerprints does not overlap each other
  completely. Hence these finger prints are of two
  different persons.

70

• Advantages
• It is the simplest method for fingerprint
  authentication.
• It is less time consuming method.
• It is comparatively easy to implement.
• It is an interactive method for recognizing
  fingerprints.
• Disadvantages
• It does not give us accurate result everytime
  because it does not relates the minutia features.
• We can only overlap the pictures having same
  dimensions.

71
Comparing slopes of ridges
72
Comparing slopes of ridges

• This method is mainly used in these days for
  fingerprint authentication. In this method we
  compare the slopes of the ridges of two images of
  fingerprints. This method is mainly based upon
  matching minutia.

73
Program for comparing slopes
74

• Program
• Step 1 Load Image
• RGB imread('fingerprint.bmp')
• imshow(RGB)
• imview(RGB)

Original image
75

• Step 2 Extract The Region Of Interest
•  
• start_row 73
• start_col 105
• cropRGB RGB(start_row134, start_col233, )
• figure, imshow(cropRGB)
• imview(RGB)
• offsetX start_col-1
• offsetY start_row-1

Cropped image
76

• Step 3 Threshold The Image
• Convert the image to black and white for
  subsequent extraction of the edge coordinates
  using bwtraceboundary routine.
• I rgb2gray(cropRGB)
• threshold graythresh(I)
• BW im2bw(I,threshold)
• BW BW
• Figure, imshow(BW)
• i2edge(BW, 'canny')
• figure, imshow(i2)
• imview(i2)

Binary image
Canny image
77

• Step 4 Find Initial Point On Each Boundary
• The bwtraceboundary routine requires that you
  specify a single point on a boundary. This point
  is used as the starting location for the boundary
  tracing process.
• dim size(i2)
• col1 40
• row1 min(find(i2(,col1)))
• row2 30
• col2 min(find(i2(row2,)))

78

• Step 5 Trace The Boundaries
• boundary1 bwtraceboundary(BW, row1, col1,
  'N', 8, 70)
• boundary2 bwtraceboundary(i2, row2, col2,
  'E', 8, 90,'counter')
• figure, imshow(RGB)
•  
•  
•  
• imview(RGB)
• plot(offsetXboundary1(,2),offsetYboundary1(,1)
  ,'g','LineWidth',2)
• plot(offsetXboundary2(,2),offsetYboundary2(,1)
  ,'g','LineWidth',2)

Marked image
79

• Step 6 Fit Lines To The Boundaries
• ab1 polyfit(boundary1(,2), boundary1(,1), 1)
• ab2 polyfit(boundary2(,2), boundary2(,1), 1)

80

• Step 7 Find The Angle Of Intersection
• Use the dot product to find the angle.
• vect1 1 ab1(1)
• vect2 1 ab2(1)
• dp dot(vect1, vect2)
• length1 sqrt(sum(vect1.2))
• length2 sqrt(sum(vect2.2))
• angle 180-acos(dp/(length1length2))180/pi

81
Result of all used command in program
82
Result

• Angle between the marked ridges of image is
  142.0786 .

83

• Advantages
• This one is more accurate than the overlapping
  method because it is based upon minutia.
• It is an interactive method for recognizing
  fingerprints.
• Disadvantages
• It is more time consuming as compared to the
  former.
• More complex program.