1. Laboratory of Image Processing Pier Luigi Mazzeo pierluigi.mazzeo@cnr.it November 4 th, 2014

2. Outline Introduction to MATLAB Image Processing with MATLAB  Simple operations with images 2

3. Reference An Introduction to Digital Image Processing with MATLAB by Alasdair MacAndrew. (download from www.ino.it/home/mazzeo/downloads/) www.ino.it/home/mazzeo/downloads/ Digital Image Processing using MATLAB. 2° Edition Rafael C. Gonzales, Richard E. Woods, Steven L. Eddins. Gatesmark Publishing. 3

4. What is MATLAB? MATLAB = Matrix Laboratory “MATLAB is a high-level language and interactive environment that enables you to perform computationally intensive tasks faster than with traditional programming languages such as C, C++ and Fortran.” (www.mathworks.com)www.mathworks.com MATLAB is an interactive, interpreted language that is designed for fast numerical matrix calculations 4

5. The MATLAB Environment MATLAB window components: Workspace > Displays all the defined variables Command Window > To execute commands in the MATLAB environment Command History > Displays record of the commands used File Editor Window > Define your functions 5

6. MATLAB Help MATLAB Help is an extremely powerful assistance to learning MATLAB Help not only contains the theoretical background, but also shows demos for implementation MATLAB Help can be opened by using the HELP pull-down menu 6

7. MATLAB Help (cont.) Any command description can be found by typing the command in the search field As shown above, the command to take square root ( sqrt ) is searched We can also utilize MATLAB Help from the command window as shown 7

8. More about the Workspace who, whos – current variables in the workspace save – save workspace variables to *.mat file load – load variables from *.mat file clear – clear workspace variables 8

9. Matrices in MATLAB Matrix is the main MATLAB data type How to build a matrix? – A=[1 2 3; 4 5 6; 7 8 9] ; – Creates matrix A of size 3 x 3 Special matrices: – zeros(n,m), ones(n,m), eye(n,m), rand(), randn() 9

10. Basic Operations on Matrices All operators in MATLAB are defined on matrices: +, -, *, /, ^, sqrt, sin, cos, etc. Element-wise operators defined with a preceding dot :.*,./,.^ size(A) – size vector sum(A) – columns sums vector sum(sum(A)) – sum of all the elements 10

11. Variable Name in Matlab Variable naming rules - must be unique in the first 63 characters - must begin with a letter - may not contain blank spaces or other types of punctuation - may contain any combination of letters, digits, and underscores - are case-sensitive - should not use Matlab keyword Pre-defined variable names pi 11

12. Logical Operators ==,, (not equal) ~=, (not) ~ find(‘condition’) – Returns indexes of A’s elements that satisfy the condition 12

13. Logical Operators (cont.) Example: >>A=[7 3 5; 6 2 1], Idx=find(A<4) A= 7 3 5 6 2 1 Idx= 3 4 6 13

14. Flow Control MATLAB has five flow control constructs: – if statement – switch statement – for loop – while loop – break statement 14

15. if IF statement condition – The general form of the IF statement is IF expression statements ELSEIF expression statements ELSE statements END 15

16. switch SWITCH – Switch among several cases based on expression The general form of SWITCH statement is: SWITCH switch_expr CASE case_expr, statement, …, statement CASE {case_expr1, case_expr2, case_expr3, …} statement, …, statement … OTHERWISE statement, …, statement END 16

17. switch (cont.) Note: – Only the statements between the matching CASE and the next CASE, OTHERWISE, or END are executed – Unlike C, the SWITCH statement does not fall through (so BREAK s are unnecessary) 17

18. for FOR repeats statements a specific number of times The general form of a FOR statement is: FOR variable=expr statements END 18

19. while WHILE repeats statements an indefinite number of times The general form of a WHILE statement is : WHILE expression statements END 19

20. Scripts and Functions There are two kinds of M-files: – Scripts, which do not accept input arguments or return output arguments. They operate on data in the workspace – Functions, which can accept input arguments and return output arguments. Internal variables are local to the function 20

21. Functions in MATLAB (cont.) Example: – A file called STAT.M: function [mean, stdev]=stat(x) %STAT Interesting statistics. n=length(x); mean=sum(x)/n; stdev=sqrt(sum((x-mean).^2)/n); – Defines a new function called STAT that calculates the mean and standard deviation of a vector. Function name and file name should be the SAME! 21

22. Visualization and Graphics plot(x,y),plot(x,sin(x)) – plot 1D function figure, figure(k) – open a new figure hold on, hold off – refreshing axis([xmin xmax ymin ymax]) – change axes title(‘figure titile’) – add title to figure mesh(x_ax,y_ax,z_mat) – view surface contour(z_mat) – view z as topo map subplot(3,1,2) – locate several plots in figure 22

23. Saving your Work save mysession % creates mysession.mat with all variables save mysession a b % save only variables a and b clear all % clear all variables clear a b % clear variables a and b load mysession % load session 23

24. Outline Introduction to MATLAB Image Processing with MATLAB  Simple operations with images 24

25. A digital image differs from a photo in that the values are all discrete. Usually they take on only integer values. A digital image can be considered as a large array of discrete dots, each of which has a brightness associated with it. These dots are called picture elements, or more simply pixels. The pixels surrounding a given pixel constitute its neighborhood A neighborhood can be characterized by its shape in the same way as a matrix: we can speak of a 3x3 neighborhood, or of a 5x7 neighborhood. Image and digital images 25

26. Pixel neighbourhood 26

27. Aspects of image processing Image Enhancement: Processing an image so that the result is more suitable for a particular application. (sharpening or deblurring an out of focus image, highlighting edges, improving image contrast, or brightening an image, removing noise). Image Restoration: This may be considered as reversing the damage done to an image by a known cause. (removing of blur caused by linear motion, removal of optical distortions). Image Segmentation: This involves subdividing an image intoconstituent parts, or isolating certain aspects of an image.(finding lines, circles, or particular shapes in an image, in an aerial photograph, identifying cars, trees, buildings, or roads. 27

28. Binary: Each pixel is just black or white. Since there are only two possible values for each pixel (0,1), we only need one bit per pixel. Grayscale: Each pixel is a shade of gray, normally from 0 (black) to 255 (white). This range means that each pixel can be represented by eight bits, or exactly one byte. Other greyscale ranges are used, but generally they are a power of 2. True Color, or RGB: Each pixel has a particular color; that color is described by the amount of red, green and blue in it. If each of these components has a range 0–255, this gives a total of 256 3 different possible colors. Such an image is a “stack” of three matrices; representing the red, green and blue values for each pixel. This means that for every pixel there correspond 3 values. Types of digital image 28

29. Binary Image 29

30. Greyscale Image 30

31. Color Image 31

32. General commands imread : Read an image figure : creates a figure on the screen. imshow(g) : which displays the matrix g as an image. impixel(i,j) : the command returns the value of the pixel (i,j) iminfo : Information about the image. 32

33. Data types 33

34. Image Information 34

35. Converting between different data type gray2ind - intensity image to index image im2bw - image to binary im2double - image to double precision im2uint8 - image to 8-bit unsigned integers im2uint16 - image to 16-bit unsigned integers ind2gray - indexed image to intensity image mat2gray - matrix to intensity image rgb2gray - RGB image to grayscale rgb2ind - RGB image to indexed image 35

36. Greyscale images can be transformed into a sequence of binary images by breaking them up into their bit-planes. We consider the grey value of each pixel of an 8-bit image as an 8-bit binary word. The 0th bit plane consists of the last bit of each grey value. Since this bit has the least effect (least significant bit plane). The 7th bit plane consists of the first bit in each value (most significant bit plane). Bit planes 36

37. Initial Image 37

38. Bit plane 0 38

39. Bit plane 4 39

40. Bit plane 7 40

41. Spatial resolution is the density of pixels over the image: the greater the spatial resolution, the more pixels are used to display the image. Halve the size of the image: It does this by taking out every other row and every other column, thus leaving only those matrix elements whose row and column indices are even. Double the size of the image: all the pixels are repeated to produce an image with the same size as the original, but with half the resolution in each direction. Spatial Resolution 41

42. Interpolation 42

43. Extrapolation 43

44. These operations act by applying a simple function y=f(x) to each gray value in the image. Simple functions include adding or subtract a constant value to each pixel: y = x±C ( imadd, imsubtract ) Multiplying each pixel by a constant: y = C·x ( immultiply, imdivide ) Complement: For a grayscale image is its photographic negative. Point Processing: Arithmetic operations 44

45. Arithmetic operations: Addition, subtraction 45

46. Arithmetic operations: Multiplication, division 46

47. Complement 47

48. Addition Image: IImage: I+50 48

49. Subtraction Image: IImage: I-80 49

50. Multiplication Image: IImage: I*3 50

51. Division Image: IImage: I/2 51

52. Complement Image: IImage: 255-I 52

53. What is the Image Processing Toolbox? The Image Processing Toolbox is a collection of functions that extend the capabilities of the MATLAB’s numeric computing environment. The toolbox supports a wide range of image processing operations, including: – Geometric operations – Neighborhood and block operations – Linear filtering and filter design – Transforms – Image analysis and enhancement – Binary image operations – Region of interest operations 53

54. Images in MATLAB MATLAB can import/export several image formats: –BMP (Microsoft Windows Bitmap) –GIF (Graphics Interchange Files) –HDF (Hierarchical Data Format) –JPEG (Joint Photographic Experts Group) –PCX (Paintbrush) –PNG (Portable Network Graphics) –TIFF (Tagged Image File Format) –XWD (X Window Dump) –raw-data and other types of image data Data types in MATLAB –Double (64-bit double-precision floating point) –Single (32-bit single-precision floating point) –Int32 (32-bit signed integer) –Int16 (16-bit signed integer) –Int8 (8-bit signed integer) –Uint32 (32-bit unsigned integer) –Uint16 (16-bit unsigned integer) –Uint8 (8-bit unsigned integer) 54

55. Images in MATLAB Binary images : {0,1} Intensity images : [0,1] or uint8, double etc. RGB images : m × n × 3 Multidimensional images: m × n × p (p is the number of layers) 55

56. Image Import and Export Read and write images in Matlab img = imread('apple.jpg'); dim = size(img); figure; imshow(img); imwrite(img, 'output.bmp', 'bmp'); Alternatives to imshow imagesc(I) imtool(I) image(I) 56

57. Images and Matrices Column 1 to 256 Row 1 to 256 o [0, 0] o [256, 256] How to build a matrix (or image)? Intensity Image: row = 256; col = 256; img = zeros(row, col); img(100:105, :) = 0.5; img(:, 100:105) = 1; figure; imshow(img); 57

58. Images and Matrices Binary Image: row = 256; col = 256; img = rand(row, col); img = round(img); figure; imshow(img); 58

59. Image Display image - create and display image object imagesc - scale and display as image imshow - display image colorbar - display colorbar getimage - get image data from axes truesize - adjust display size of image zoom - zoom in and zoom out of 2D plot 59

60. Image Conversion gray2ind - intensity image to index image im2bw - image to binary im2double - image to double precision im2uint8 - image to 8-bit unsigned integers im2uint16 - image to 16-bit unsigned integers ind2gray - indexed image to intensity image mat2gray - matrix to intensity image rgb2gray - RGB image to grayscale rgb2ind - RGB image to indexed image 60

61. Image Operations RGB image to gray image Image resize Image crop Image rotate Image histogram Image histogram equalization Image DCT/IDCT Convolution 61

62. Outline Introduction to MATLAB –Basics & Examples Image Processing with MATLAB –Basics & Examples 62

63. Examples working with Images (2/3) Blending two images 63

64. Examples working with Images (3/3) Sobel descriptor to detect object edge 64

65. Performance Issues The idea: MATLAB is –very fast on vector and matrix operations –Correspondingly slow with loops Try to avoid loops Try to vectorize your code http://www.mathworks.com/support/tech- notes/1100/1109.html 65

66. Vectorize Loops Example –Given image matrices, A and B, of the same size (540*380), blend these two images apple = imread(‘apple.jpg'); orange = imread(‘orange.jpg’); Poor Style % measure performance using stopwatch timer tic for i = 1 : size(apple, 1) for j = 1 : size(apple, 2) for k = 1 : size(apple, 3) output(i, j, k) = (apple(i, j, k) + orange(i, j, k))/2; end toc Elapsed time is 0.138116 seconds0.138116 66

67. Vectorize Loops (cont.) Example –Given image matrices, A and B, of the same size (600*400), blend these two images apple = imread(‘apple.jpg'); orange = imread(‘orange.jpg’); Better Style tic % measure performance using stopwatch timer Output = (apple + orange)/2; toc Elapsed time is 0.099802 seconds0.099802 Computation is faster! 67