Download presentation
Presentation is loading. Please wait.
Published byAmanda Garrison Modified over 9 years ago
1
1 Regions and Binary Images Hao Jiang Computer Science Department Sept. 24, 2009
2
Figure Ground Separation 2
3
Brightness Thresholding 3
4
= - Thresholding Given a grayscale image or an intermediate matrix threshold to create a binary output. Example: background subtraction Looking for pixels that differ significantly from the “empty” background. fg_pix = find(diff > t); Slide from Kristen Grauman
5
Thresholding Given a grayscale image or an intermediate matrix threshold to create a binary output. Example: color-based detection Looking for pixels within a certain hue range. fg_pix = find(hue > t1 & hue < t2); Slide from Kristen Grauman
6
More Binary Images 6 Slide from Kristen Grauman
7
Issues How to demarcate multiple regions of interest? Count objects Compute further features per object What to do with “noisy” binary outputs? Holes Extra small fragments Slide from Kristen Grauman
8
Find Connected Regions 8
9
9 Our target in this image is the largest blob.
10
Connected Components Identify distinct regions of “connected pixels” Shapiro and Stockman
11
Pixel Neighbors 11 4 neighboring pixels of the blue pixel
12
Pixel Neighbors 12 8 neighboring pixels of the blue pixel
13
Recursive Method 13 label = 2 for i = 1 to rows for j = 1 to cols if I(i, j) == 1 labelConnectedRegion(i, j, label) label ++; end
14
Recursive Method 14 function labelConnectedRegion(int i, int j, int label) if (i,j) is labeled or background or out of boundary return I(i,j)=label for (m,n) belongs to neighbors of (i,j) labelConnectedRegion(m,n,label) end
15
Two Pass Method We check each pixel from left to right and up to bottom If a pixel has no left and up foreground neighbor, we assign a new label to the pixel If a pixel has only one left or up foreground neighbor, we assign the label of the neighbor to the pixel If a pixel has both left and up foreground neighbors, and their labels are the same, we assign the label of the neighbor to the pixel If a pixel has both left and up foreground neighbors, and their labels are the different, we assign the label of the left neighbor to the pixel and use the up-left label pair to update the equivalency table. We go through another pass to replace the labels to corresponding labels in the equivalency table 15
16
Example 16 11111 1111 1111 11111 11111 22 33 44 55 66 77 88 99 Image Label equivalence table
17
Example 17 21111 1111 1111 11111 11111 22 33 44 55 66 77 88 99 Image Label equivalence table
18
Example 18 22111 1111 1111 11111 11111 22 33 44 55 66 77 88 99 Image Label equivalence table
19
Example 19 22311 1111 1111 11111 11111 22 33 44 55 66 77 88 99 Image Label equivalence table
20
Example 20 22331 1111 1111 11111 11111 22 33 44 55 66 77 88 99 Image Label equivalence table
21
Example 21 22333 1111 1111 11111 11111 22 33 44 55 66 77 88 99 Image Label equivalence table
22
Example 22 22333 2111 1111 11111 11111 22 33 44 55 66 77 88 99 Image Label equivalence table
23
Example 23 22333 2211 1111 11111 11111 22 33 44 55 66 77 88 99 Image Label equivalence table
24
Example 24 22333 2221 1111 11111 11111 22 33 44 55 66 77 88 99 Image Label equivalence table
25
Example 25 22333 2222 1111 11111 11111 23 33 44 55 66 77 88 99 Image Label equivalence table (2,3)
26
Example 26 22333 2222 2111 11111 11111 23 33 44 55 66 77 88 99 Image Label equivalence table
27
Example 27 22333 2222 2211 11111 11111 23 33 44 55 66 77 88 99 Image Label equivalence table
28
Example 28 22333 2222 2241 11111 11111 23 33 44 55 66 77 88 99 Image Label equivalence table
29
Example 29 22333 2222 2244 11111 11111 23 33 44 55 66 77 88 99 Image Label equivalence table
30
Example 30 22333 2222 2244 51111 11111 23 33 44 55 66 77 88 99 Image Label equivalence table
31
Example 31 22333 2222 2244 55111 11111 23 33 44 53 66 77 88 99 Image Label equivalence table (5,2)
32
Example 32 22333 2222 2244 55211 11111 23 33 44 53 66 77 88 99 Image Label equivalence table
33
Example 33 22333 2222 2244 55241 11111 23 33 44 53 66 77 88 99 Image Label equivalence table
34
Example 34 22333 2222 2244 55244 11111 23 33 44 53 66 77 88 99 Image Label equivalence table
35
Example 35 22333 2222 2244 55244 51111 23 33 44 53 66 77 88 99 Image Label equivalence table
36
Example 36 22333 2222 2244 55244 55111 23 33 44 53 66 77 88 99 Image Label equivalence table
37
Example 37 22333 2222 2244 55244 55511 23 33 44 53 66 77 88 99 Image Label equivalence table
38
Example 38 22333 2222 2244 55244 55551 23 33 44 53 66 77 88 99 Image Label equivalence table (5,2)
39
Example 39 22333 2222 2244 55244 55554 23 33 44 53 66 77 88 99 Image Label equivalence table
40
Example 40 33333 3333 3344 33344 33334 23 33 44 53 66 77 88 99 Image Label equivalence table
41
Morphology Operations Define At = { p + t | p is a point in A} Erosion 41 T = { t | At belongs to S} A S Erosion(S, A)
42
Morphology Operations Define At = { p + t | p is a point in A} Erosion 42 T = { t | At belongs to S} S What if A is
43
Morphology Operations Define At = { p + t | p is a point in A} Dilation 43 T = Union of At and S for all t in S S What if A is Erosion(S, A)
44
Morphology Operations Define At = { p + t | p is a point in A} Dilation 44 T = Union of At and S for all t in S S What if A is
45
Morphology Operations Define At = { p + t | p is a point in A} Dilation 45 T = Union of At and S for all t in S What if A is
46
Opening Erode, then dilate Remove small objects, keep original shape Before openingAfter opening Slide from Kristen Grauman
47
Closing Dilate, then erode Fill holes, but keep original shape Before closingAfter closing Slide from Kristen Grauman
48
Morphology Operators on Grayscale Images Dilation and erosion typically performed on binary images. If image is grayscale: for dilation take the neighborhood max, for erosion take the min. originaldilated eroded Slide from Kristen Grauman
49
Matlab Create structure element se = strel(‘disk’, radius); Erosion imerode(image, se); Dilation imdilate(image, se); Opening imopen(image, se); Closing imclose(image, se); More possibilities bwmorph(image, ‘skel’); 49
50
Figure Ground Separation 50
51
Brightness Thresholding 51
52
Opening 52
53
Find the Largest Connected Region 53
54
Example Using Binary Image Analysis: segmentation of a liver Slide credit: Li Shen Slide from Kristen Grauman
55
Example Using Binary Image Analysis: Bg subtraction + blob detection … Slide from Kristen Grauman
56
University of Southern California http://iris.usc.edu/~icohen/projects/vace/detection.htm Example Using Binary Image Analysis: Bg subtraction + blob detection Slide from Kristen Grauman
Similar presentations
© 2024 SlidePlayer.com. Inc.
All rights reserved.