Computer and Robot Vision I

1. Computer and Robot Vision I Chapter 10 Image Segmentation Presented by:陳雲濤 0921192337 r03525056@ntu.edu.tw 指導教授: 傅楸善 博士 Digital Camera and Computer Vision Laboratory Department of Computer Science and Information Engineering National Taiwan University, Taipei, Taiwan, R.O.C.

2. 10.0 關於投影片 • Online slide: https://goo.gl/xgYW9X • 共同編輯筆記吧 :D DC & CV Lab. NTU CSIE

3. 10.1 Introduction • image segmentation: • partition of image into set of non-overlapping regions • union of segmented regions is the entire image • segmentation purpose: • to decompose image into meaningful parts to application • segmentation based on valleys in gray level histogram into regions DC & CV Lab. NTU CSIE

4. 10.1 Introduction (cont’) DC & CV Lab. NTU CSIE

5. DC & CV Lab. NTU CSIE

6. DC & CV Lab. NTU CSIE

7. 10.1 Introduction (cont’) • rules for general segmentation procedures • region uniform, homogeneous w.r.t. characteristic e.g. gray level, texture • region interiors simple and without many small holes • adjacent regions with significantly different values on characteristic • boundaries simple not ragged spatially accurate DC & CV Lab. NTU CSIE

8. 10.1 Introduction (cont’) • Clustering: • process of partitioning set of pattern vectors into clusters • set of points in Euclidean measurement space separated into 3 clusters • In some sense close to one another DC & CV Lab. NTU CSIE

9. 10.1 Introduction (cont’) DC & CV Lab. NTU CSIE

10. 10.1 Introduction (cont’) • no full theory of clustering • no full theory of image segmentation • image segmentation techniques: ad hoc, different in emphasis and compromise DC & CV Lab. NTU CSIE

11. Joke DC & CV Lab. NTU CSIE

12. 10.2 Measurement-Space-Guided Spatial Clustering • The technique of measurement-space-guided spatial clustering for image segmentation uses the measurement-space-clustering process to define a partition in measurement space. DC & CV Lab. NTU CSIE

13. 10.2 Measurement-Space-Guided Spatial Clustering (cont’) • histogram mode seeking: • a measurement-space-clustering process • homogeneous objects as clusters in histogram • one pass, the least computation time DC & CV Lab. NTU CSIE

14. 10.2 Measurement-Space-Guided Spatial Clustering (cont’) • enlarged image of a polished mineral ore section DC & CV Lab. NTU CSIE

15. 10.2 Measurement-Space-Guided Spatial Clustering (cont’) • 3 nonoverlapping modes: • black holes • Pyrorhotite • pyrite DC & CV Lab. NTU CSIE

16. 10.2 Measurement-Space-Guided Spatial Clustering (cont’) DC & CV Lab. NTU CSIE

17. 10.2 Measurement-Space-Guided Spatial Clustering (cont’) • 2 valleys in histogram is a virtually perfect (meaningful) segmentation DC & CV Lab. NTU CSIE

18. 10.2 Measurement-Space-Guided Spatial Clustering (cont’) • 原圖 • 切割後的結果 DC & CV Lab. NTU CSIE

19. 10.2 Measurement-Space-Guided Spatial Clustering (cont’) • example image not ideal for measurement-space-clustering image segmentation DC & CV Lab. NTU CSIE

20. 10.2 Measurement-Space-Guided Spatial Clustering (cont’) • histogram with three modes and two valleys DC & CV Lab. NTU CSIE

21. 10.2 Measurement-Space-Guided Spatial Clustering (cont’) • Undesirable: many border regions show up as dark segments DC & CV Lab. NTU CSIE

22. 10.2 比較 DC & CV Lab. NTU CSIE

23. 10.2 Measurement-Space-Guided Spatial Clustering (cont’) • segmentation into homogeneous regions: not necessarily good solution DC & CV Lab. NTU CSIE

24. joke DC & CV Lab. NTU CSIE

25. 10.2 Measurement-Space-Guided Spatial Clustering (cont’) • diagram of an F-15 bulkhead DC & CV Lab. NTU CSIE

26. 10.2 Measurement-Space-Guided Spatial Clustering (cont’) • image of a section of the F-15 bulkhead DC & CV Lab. NTU CSIE

27. 10.2 Measurement-Space-Guided Spatial Clustering (cont’) • Histogram of the image DC & CV Lab. NTU CSIE

28. 10.2 Measurement-Space-Guided Spatial Clustering (cont’) • Five clusters: bad spatial continuation, boundaries noisy and busy DC & CV Lab. NTU CSIE

29. 10.2 Measurement-Space-Guided Spatial Clustering (cont’) • three clusters: less boundary noise, but much less detail DC & CV Lab. NTU CSIE

30. 10.2 Measurement-Space-Guided Spatial Clustering (cont’) • recursive histogram-directed spatial clustering DC & CV Lab. NTU CSIE

31. DC & CV Lab. NTU CSIE

32. 10.2 Measurement-Space-Guided Spatial Clustering (cont’) • applied to the bulkhead image DC & CV Lab. NTU CSIE

33. 10.2 Measurement-Space-Guided Spatial Clustering (cont’) • performing morphological opening with 3 x 3 square structuring element DC & CV Lab. NTU CSIE

34. 10.2 Measurement-Space-Guided Spatial Clustering (cont’) • tiny regions removed, but several long, thin regions lost DC & CV Lab. NTU CSIE

35. 10.2 Measurement-Space-Guided Spatial Clustering (cont’) • a color image DC & CV Lab. NTU CSIE

36. 10.2 Measurement-Space-Guided Spatial Clustering (cont’) • recursive histogram-directed spatial clustering using R,G,B bands and other DC & CV Lab. NTU CSIE

37. joke DC & CV Lab. NTU CSIE

38. 10.2.1 Thresholding • Kohler denotes the set E(T) of edges detected by a threshold T to be the set of all pairs of neighboring pixels one of whose gray level intensities is less than or equal to T and one of whose gray level intensities is greater than T DC & CV Lab. NTU CSIE

39. 10.2.1 Thresholding (cont’) • 1. pixels and are neighbors • 2.min max DC & CV Lab. NTU CSIE

40. 10.2.1 Thresholding (cont’) • The total contrast C(T) of edges detected by threshold T is given by DC & CV Lab. NTU CSIE

41. 10.2.1 Thresholding (cont’) • The average contrast of all edges detected by threshold T: • The best threshold Tb is determined by the value that maximizes DC & CV Lab. NTU CSIE

42. 10.2.1 Thresholding (cont’) • Example • T = 50 • [45, 110] • [33, 88] • [15, 65] • [45, 115] • [0, 225] DC & CV Lab. NTU CSIE

43. 10.2.1 Thresholding (cont’) • Example • T = 50 • [5, 60] • [17, 33] • [35, 15] • [5, 65] • [50, 175] DC & CV Lab. NTU CSIE

44. 10.2.1 Thresholding (cont’) • Example • T = 50 • [5, 60] • [17, 33] • [35, 15] • [5, 65] • [50, 175] The average contrast of all edges (5+17+15+5+50) / 5 = 18.4 DC & CV Lab. NTU CSIE

45. 10.2.1 Thresholding (cont’) • Example • T = 50 • [5, 60] • [17, 33] • [35, 15] • [5, 65] • [50, 175] The average contrast of all edges (5+17+15+5+50) / 5 = 18.4 The best threshold Tb is determined by the maximizes DC & CV Lab. NTU CSIE

46. 10.2.1 Thresholding (cont’) • approach for segmenting white blob against dark background • pixel with small gradient: not likely to be an edge DC & CV Lab. NTU CSIE

47. 10.2.1 Thresholding (cont’) • if not an edge, then either dark background pixel or bright blob pixel • histogram of small gradient pixels: bimodal DC & CV Lab. NTU CSIE

48. 10.2.1 Thresholding (cont’) • pixels with small gradients: valley between two modes: threshold point DC & CV Lab. NTU CSIE

49. 10.2.1 Thresholding (cont’) • FLIR (Forward Looking Infra-Red) image from NATO (North Atlantic Treaty Organization) database DC & CV Lab. NTU CSIE

50. 10.2.1 Thresholding (cont’) • thresholded at gray level intensity 159 and 190 DC & CV Lab. NTU CSIE