Introduction to Computer and Human Vision

1. Introduction to Computer and Human Vision Shimon Ullman, Ronen Basri, Michal Irani Assistants: Tal Hassner <hassner@wisdom.weizmann.ac.il> Eli Shechtman <elishe @wisdom.weizmann.ac.il>

2. Misc... • Course website: www.wisdom.weizmann.ac.il/~hassner/cv0203 • To be added to course mailing-list: send email to <hassner@wisdom….> • Other recommended courses (for credit): - Basic Topics - Statistical Machine Learning • Vision & Robotics Seminar (not for credit): Thursdays at 11:00-12:00 (Ziskind 1) send email <leah@wisdom…>ask to be added to “seminar13”mailing list

3. Applications: - Manufacturing and inspection; QA - Robot navigation - Autonomous vehicles - Guiding tools for blind - Security and monitoring - Object/face recognition; OCR. - Medical Applications - Visualization; NVS - Visual communication - Digital libraries and video search - Video manipulation and editing • How is an image formed? (geometry and photometry) • What kind of operations can we apply to images? • What do images tell us about the world? (analysis & interpretation)

4. Tentative Schedule Lessons 1-3 (Michal): Basic Image Processing Lessons 4-6 (Ronen): Stereo and Structure from Motion Lessons 7-9 (Michal): Motion and video analysis Lesson 10 (Ronen): Image Segmentation Lesson 11 (Ronen): Photometry Lesson 12 (Shimon): Object recognition Lessons 13-14 (Shimon): Human Vision • 3 programming exercises (MATLAB) -- CAN SUBMIT IN PAIRS • 3-4 theoretical exercises -- MUST SUBMIT INDIVIDUALLY • EXAM

5. Digital Images World Camera Digitizer Digital Image Image Formation: (i) What determines where the image of a 3D point appears on the 2D image? (ii) What determines how bright that image point is? (iii) How is a digital image represented? (iv) Some simple operations on 2D images? today

6. 0 10 10 15 50 70 80 0 0 100 120 125 130 130 0 35 100 150 150 80 50 0 15 70 100 10 20 20 0 15 70 0 0 0 15 5 15 50 120 110 130 110 5 10 20 50 50 20 250 PIXEL (picture element) Typically: 0 = black 255 = white Digital Images World Camera Digitizer Digital Image

7. 64 60 69 100 149 151 176 182 179 65 62 68 97 145 148 175 183 181 65 66 70 95 142 146 176 185 184 66 66 68 90 135 140 172 184 184 66 64 64 84 129 134 168 181 182 59 63 62 88 130 128 166 185 180 60 62 60 85 127 125 163 183 178 62 62 58 81 122 120 160 181 176 63 64 58 78 118 117 159 180 176

8. Grayscale Image x = 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 210 209 204 202 197 247 143 71 64 80 84 54 54 57 58 206 196 203 197 195 210 207 56 63 58 53 53 61 62 51 201 207 192 201 198 213 156 69 65 57 55 52 53 60 50 216 206 211 193 202 207 208 57 69 60 55 77 49 62 61 221 206 211 194 196 197 220 56 63 60 55 46 97 58 106 209 214 224 199 194 193 204 173 64 60 59 51 62 56 48 204 212 213 208 191 190 191 214 60 62 66 76 51 49 55 214 215 215 207 208 180 172 188 69 72 55 49 56 52 56 209 205 214 205 204 196 187 196 86 62 66 87 57 60 48 208 209 205 203 202 186 174 185 149 71 63 55 55 45 56 207 210 211 199 217 194 183 177 209 90 62 64 52 93 52 208 205 209 209 197 194 183 187 187 239 58 68 61 51 56 204 206 203 209 195 203 188 185 183 221 75 61 58 60 60 200 203 199 236 188 197 183 190 183 196 122 63 58 64 66 205 210 202 203 199 197 196 181 173 186 105 62 57 64 63 y =

9. Three types of images: • Gray-scale images I(x,y)  [0..255] • Binary images I(x,y)  {0 , 1} • Color images IR(x,y) IG(x,y) IB(x,y)

10. Color Image

11. Effects of down-sampling (reducing number of pixels) 128 x 128 64 x 64 32 x 32 16 x 16 8 x 8 4 x 4

12. Effects of reducing number of gray levels 256 gray levels (8 bits/pixel) 16 gray levels (4 bits/pixel) 8 gray levels (3 bits/pixel) 4 gray levels (2 bits/pixel) 2 gray levels (1 bit/pixel) BINARY IMAGE

13. Continuous probability density function: The Image Histogram Occurrence (# of pixels) Gray Level Histogram = The gray-level distribution: H(k) = #pixels with gray-level k Normalized histogram: Hnorm(k)=H(k)/N (N = # pixels in the image)

14. The Image Histogram (Cont.) PI(k) 1 k PI(k) 1 0.5 k PI(k) 0.1 k

15. PI(k) 0.1 k 0.5 PI(k) 0.1 k Histogram Stretching

16. k k Histogram Equalization k k

17. Histogram Equalization Original Equalized

18. Histogram Equalization 3000 3000 2500 2500 2000 2000 1500 1500 1000 1000 500 500 0 0 0 50 100 150 200 250 0 50 100 150 200 250 Original Equalized

19. Histogram Specification Transforms an image so that its histogram matches that of another image (e.g., for comparing two images of the same scene acquired under different lighting condition) Aa Ab k k

20. Image Enhancement 1) Gray value (histogram) Domain 2) Spatial Domain 3) Frequency Domain - Histogram stretching - Histogram equalization - Histogram specification - Gamma correction etc... noisy image(salt & pepper noise)

21. g(x,y) = 1/M S f(n,m) (n,m) inS Spatial Operations Replace center pixel with average/median level: (averaging mask; weighted mask; median filter…) Examples of neighborhoods S: 3 x 3 5 x 5 S = neighborhood of pixel (x,y) M = number of pixels in neighborhood S e.g.,

22. Noise Cleaning Salt & Pepper Noise 3 X 3 Average 5 X 5 Average 7 X 7 Average Median

23. Noise Cleaning Salt & Pepper Noise 3 X 3 Average 5 X 5 Average 7 X 7 Average Median

24. Other spatial filters Are strong brightness variations always noise…?

25. Edge Detection

28. Edge Types Line Edge gray value x edge edge Step Edge gray value x edge edge

29. Edge Detection by Differentiation gray value 1D image f(x) x 1st derivative f'(x) threshold |f'(x)| Edge Pixels: |f'(x)| > Threshold

30. Original image x derivative y derivative Gradient magnitude

31. Edge Detection Image Vertical edges Horizontal edges

32. Edge Detection Image

33. Image Sharpening Blurry Image Laplacian Sharpened Image Also Laplacian; Zero-crossings; Edge sharpening; etc….

34. The End... • Exercise#1: Noise Cleaning -- on course website (+ Matlab tutorial) DUE: Nov. 10 (in 2 weeks) • Course mailing list: Send email to <hassner@wisdom….> • Vision & Robotics Seminar: send email <leah@wisdom…> ask to be added to “seminar13” mailing list

36. Generated Mosaic image Panoramic Mosaic Image Original video clip

37. Video Removal Original Original Outliers Synthesized

38. Image Segmentation

39. Image Segmentation

40. Photometric Stereo

41. Photometric Stereo