Operations in Digital Image Processing + Convolution by Example
5 likes2,468 views
The document focuses on operations in digital image processing, emphasizing point and group operations, along with convolution techniques. It includes step-by-step examples of arithmetic and convolution operations applied to image signals. The content is structured to provide a comprehensive understanding of how these operations manipulate image data using mathematical principles.
Operations in Digital Image Processing + Convolution by Example
- 1. Operations in Digital Image Processing + Convolution by Example MENOUFIA UNIVERSITY FACULTY OF COMPUTERS AND INFORMATION ALL DEPARTMENTS المنوفية جامعة والمعلومات الحاسبات كلية جميعاألقسام المنوفية جامعة Ahmed Fawzy Gad [email protected]
- 3. Point Operations 58 3 213 81 78 185 87 32 27 11 70 66 60 2 19 61 91 129 89 38 14 7 58 14 42 Arithmetic Operations + - * / Add 2
- 4. Point Operations 58 3 213 81 78 185 87 32 27 11 70 66 60 2 19 61 91 129 89 38 14 7 58 14 42 Arithmetic Operations + - * / Add 2
- 5. Point Operations 60 3 213 81 78 185 87 32 27 11 70 66 60 2 19 61 91 129 89 38 14 7 58 14 42 Arithmetic Operations + - * / Add 2
- 6. Point Operations 60 3 213 81 78 185 87 32 27 11 70 66 60 2 19 61 91 129 89 38 14 7 58 14 42 Arithmetic Operations + - * / Add 2
- 7. Point Operations 60 5 213 81 78 185 87 32 27 11 70 66 60 2 19 61 91 129 89 38 14 7 58 14 42 Arithmetic Operations + - * / Add 2
- 8. Point Operations 60 5 213 81 78 185 87 32 27 11 70 66 60 2 19 61 91 129 89 38 14 7 58 14 42 Arithmetic Operations + - * / Add 2
- 9. Point Operations 60 5 215 81 78 185 87 32 27 11 70 66 60 2 19 61 91 129 89 38 14 7 58 14 42 Arithmetic Operations + - * / Add 2
- 10. Point Operations 60 5 215 83 80 187 89 34 29 13 72 68 62 4 21 63 93 131 91 40 16 9 60 16 44 Arithmetic Operations + - * / Add 2
- 11. Group Operations 58 3 213 81 78 185 87 32 27 11 70 66 60 2 19 61 91 129 89 38 14 7 58 14 42Mode Median Mean
- 12. Group Operations 58 3 213 81 78 185 87 32 27 11 70 66 60 2 19 61 91 129 89 38 14 7 58 14 42Mode Median Mean
- 13. Group Operations 58 3 213 81 78 185 87 32 27 11 70 66 60 2 19 61 91 129 89 38 14 7 58 14 42Mode Median Mean 58 + 3 + 213 + 185 + 87 + 32 + 70 + 66 + 60 9 = 𝟖𝟔
- 14. Group Operations – Template Operations 1 Operation 2 Template Mode Median Mean 7x7 5x5 3x3 ODD Why odd?
- 15. Group Operations 58 3 213 81 78 185 87 32 27 11 70 66 60 2 19 61 91 129 89 38 14 7 58 14 42Mode Median Mean 58 + 3 + 213 + 185 + 87 + 32 + 70 + 66 + 60 9 = 𝟖𝟔 Put the results in the center.
- 16. Group Operations 58 3 213 81 78 185 87 32 27 11 70 66 60 2 19 61 91 129 89 38 14 7 58 14 42Mode Median Mean 58 + 3 + 213 + 185 + 87 + 32 + 70 + 66 + 60 9 = 𝟖𝟔 Put the results in the center. 𝟖𝟔
- 17. Group Operations – Even Template Size 58 3 213 81 78 185 87 32 27 11 70 66 60 2 19 61 91 129 89 38 14 7 58 14 42Mode Median Mean 58 + 3 + 185 + 87 4 = 𝟖𝟑. 𝟐𝟓 ≅ 𝟖𝟒 Even template sizes has no center. E.g. 2x3, 2x2, 5x6, … 𝟖𝟒
- 18. Group Operations – Even Template Size 58 3 213 81 78 185 87 32 27 11 70 66 60 2 19 61 91 129 89 38 14 7 58 14 42Mode Median Mean Even template sizes has no center. E.g. 2x3, 2x2, 5x6, … 𝟖𝟒 ??? 58 + 3 + 185 + 87 4 = 𝟖𝟑. 𝟐𝟓 ≅ 𝟖𝟒
- 19. What is Convolution? • It is a mathematical way to combine two signals to generate a third signal. Convolution is not limited on digital image processing and it is a broad term that works on signals. Input Signal Impulse Response Output Signal System Convolution
- 20. What is Convolution? • It is a mathematical way to combine two signals to generate a third signal. Convolution is not limited on digital image processing and it is a broad term that works on signals. Input Signal Impulse Response Output Signal System Convolution 58 3 213 81 78 185 87 32 27 11 70 66 60 2 19 61 91 129 89 38 14 7 58 14 42
- 21. Convolution for 2D Image Signal 1 3 4 3 10 2 7 4 1 11 6 2 5 2 5 13 6 8 9 1 2 7 12 14 8Template – 3x3 2 1 -4 3 2 5 -1 8 1
- 22. Convolution for 2D Image Signal Template – 3x3 2 1 -4 3 2 5 -1 8 1 1 3 4 3 10 2 7 4 1 11 6 2 5 2 5 13 6 8 9 1 2 7 12 14 8
- 23. Convolution for 2D Image Signal Template – 3x3 2 1 -4 3 2 5 -1 8 1 1 3 4 3 10 2 7 4 1 11 6 2 5 2 5 13 6 8 9 1 2 7 12 14 8
- 24. Convolution for 2D Image Signal 1 3 4 3 10 2 44 4 1 11 6 2 5 2 5 13 6 8 9 1 2 7 12 14 8 2 ∗ 1 + 1 ∗ 3 − 4 ∗ 4 + 3 ∗ 2 + 2 ∗ 7 + 5 ∗ 4 − 1 ∗ 6 + 8 ∗ 2 + 1 ∗ 5 = 𝟒𝟒
- 25. 1 3 4 3 10 2 44 4 1 11 6 2 5 2 5 13 6 8 9 1 2 7 12 14 8 Convolution for 2D Image Signal 2 1 -4 3 2 5 -1 8 1
- 26. Convolution for 2D Image Signal 2 ∗ 3 + 1 ∗ 4 − 4 ∗ 3 + 3 ∗ 7 + 2 ∗ 4 + 5 ∗ 1 − 1 ∗ 2 + 8 ∗ 5 + 1 ∗ 2 = 𝟕𝟐 1 3 4 3 10 2 44 72 1 11 6 2 5 2 5 13 6 8 9 1 2 7 12 14 8
- 27. 1 3 4 3 10 2 44 72 1 11 6 2 5 2 5 13 6 8 9 1 2 7 12 14 8 Convolution for 2D Image Signal 2 1 -4 3 2 5 -1 8 1
- 28. 1 3 4 3 10 2 44 72 2 11 6 2 5 2 5 13 6 8 9 1 2 7 12 14 8 Convolution for 2D Image Signal 2 ∗ 4 + 1 ∗ 3 − 4 ∗ 10 + 3 ∗ 4 + 2 ∗ 1 + 5 ∗ 11 − 1 ∗ 5 + 8 ∗ 2 + 1 ∗ 5 = 𝟐
- 29. 1 3 4 3 10 2 44 72 2 11 6 2 5 2 5 13 6 8 9 1 2 7 12 14 8 Convolution for 2D Image Signal 2 1 -4 3 2 5 -1 8 1
- 30. 1 3 4 3 10 2 44 72 2 11 6 2 5 2 5 13 6 8 9 1 2 7 12 14 8 Convolution for 2D Image Signal 2 1 -4 3 2 5 -1 8 1
- 31. 1 3 4 3 10 2 44 72 2 82 6 2 5 2 5 13 6 8 9 1 2 7 12 14 8 Convolution for 2D Image Signal 2 ∗ 3 + 1 ∗ 10 − 4 ∗ 0 + 3 ∗ 2 + 2 ∗ 11 + 5 ∗ 0 − 1 ∗ 2 + 8 ∗ 5 + 1 ∗ 0 = 𝟖𝟐
- 32. 1 3 4 3 10 2 44 72 2 82 6 2 5 2 5 13 6 8 9 1 2 7 12 14 8 Convolution for 2D Image Signal
- 33. 1 3 4 3 10 2 44 72 2 82 6 2 5 2 5 13 6 8 9 1 2 7 12 14 8 Convolution for 2D Image Signal 2 1 -4 3 2 5 -1 8 1
- 34. 1 3 4 3 10 2 44 72 2 82 6 2 5 2 5 13 6 8 9 1 2 80 12 14 8 Convolution for 2D Image Signal 2 ∗ 13 + 1 ∗ 6 − 4 ∗ 8 + 3 ∗ 2 + 2 ∗ 7 + 5 ∗ 12 − 1 ∗ 0 + 8 ∗ 0 + 1 ∗ 0 = 𝟖𝟎
- 35. 1 3 4 3 10 2 44 72 2 82 6 2 5 2 5 13 6 8 9 1 2 80 12 14 8 Convolution for 2D Image Signal 2 1 -4 3 2 5 -1 8 1
- 36. 1 3 4 3 10 2 44 72 2 82 6 2 5 2 5 13 6 8 9 1 2 80 12 14 77 Convolution for 2D Image Signal 2 ∗ 9 + 1 ∗ 1 − 4 ∗ 0 + 3 ∗ 14 + 2 ∗ 8 + 5 ∗ 0 − 1 ∗ 0 + 8 ∗ 0 + 1 ∗ 0 = 𝟕𝟕
- 37. Convolution for 2D Signal •Continue. 1 3 4 3 10 2 44 72 2 82 6 2 5 2 5 13 6 8 9 1 2 80 12 14 77
- 38. Convolution for 2D Signal •Template for Mean.𝟏 𝟗 𝟏 𝟗 𝟏 𝟗 𝟏 𝟗 𝟏 𝟗 𝟏 𝟗 𝟏 𝟗 𝟏 𝟗 𝟏 𝟗 58 3 213 81 78 185 87 32 27 11 70 66 60 2 19 61 91 129 89 38 14 7 58 14 42 𝟏 𝟗 ∗ 58 + 𝟏 𝟗 ∗ 3 + 𝟏 𝟗 ∗ 213 + 𝟏 𝟗 ∗ 185 + 𝟏 𝟗 ∗ 87 + 𝟏 𝟗 ∗ 32 + 𝟏 𝟗 ∗ 70 + 𝟏 𝟗 ∗ 66 + 𝟏 𝟗 ∗ 60 = 86