Image Processing and Computer Vision - PowerPoint PPT Presentation

About This Presentation
Title:

Image Processing and Computer Vision

Description:

Image Processing and Computer Vision Edge Detection & Generalized Hough Transform – PowerPoint PPT presentation

Number of Views:141
Avg rating:3.0/5.0
Slides: 25
Provided by: Kedkarn
Category:

less

Transcript and Presenter's Notes

Title: Image Processing and Computer Vision


1
Image Processing and Computer Vision
  • Edge Detection
  • Generalized Hough Transform

2
Edge Detection
Contour edge
  • Edge ?? 2 ????
  • Contour edge
  • Texture edge

Texture edge
3
Edge Detection
  • ???????? Edge ????????????? pixel ?????????????
    ?????????????????????? intensity ??? ? ???? ???
    0-255 ??????? 255-0

4
Edge Filter (Robert Operator)
5
Example (Robert operator)
Gx
Gray scale Image
Gy
G
6
Edge Filter (Prewitt Operator)
7
Edge Filter (Sobel Operator)
8
Edge Filter (Canny Operator)
  • d. Canny operator
  • Ii,j image
  • Convolution image by Gaussian Filter we will
    get
  • Si,j Gi,j ? Ii,j
  • ? spread of gaussian (??????????????? gaussian
    filter)
  • ?????????? filter ???????
  • Pi,j (Si,j1 Si,j Si1,j1
    Si1,j) / 2
  • Qi,j (Si,j Si1,j Si,j1
    Si1,j1) /2

9
Edge Filter (Canny Operator)
  • -Edge pixel
  • Mi,j ? Pi,j2 Qi,j2
  • - Each ? in edge pixel
  • ?i,j arctan(Qi,j, Pi,j)

10
Sample Image
11
Generalized Hough Transform
  • ?????????????????? ? ?? ???????????????? (Target
    Image)
  • ?????????????? 2 ???????? ? ???
  • ??????????????????????? ? ??? Template
    shape???????
  • ????????????????? (shape) ????????????????(Target
    image)

12
Learn Shape
  • ?????????? Shape ?????????? Template ??????????
  • ?????????????????????????????????
  • ?????? pixel ??????? edge ??????
  • ???????????????????????? ????????????
  • ??????? ? ??????
  • ? ????????????????? X
  • ? ??? gradient ??????????????
  • edge detector
  • r ????????????????????????????
  • ???????????

13
GHT R-Table
????? R-Table ?????? ??????????????????? ?
14
Search for Shapes
Target Image
15
Search for Shape
  • Algorithm ?????? search ?????????????(Target)
  • 1. ?????????? ?????????????????
    ??????????????? ? (gradient direction)
  • ???
  • 2. ????????? ? edge pixel ?????? ? ???????????? 1
    ??????????????????
  • ? ??? r ?????????????? R-Table
  • 3. ???????????? ? ??? r ???????????????????
    R-Table ??? ? ???? ????????
  • ????????? ? ??? r ??????
  • xc x rcos(?)
  • yc y rsin(?)
  • 4. ????? Vote xc ??? yc ????????????????????
    Maximum vote
  • ??? xc ??? yc ????????????????????????????????
    ????????????? template
  • ????????? target

16
Search for Shapes
17
Case when Image Scale or Rotate
Scale factor S Rotation factor ? ??????????
Array ???? 4 ???? ???????????????? Vote ??????
Arrayxc, yc, S, ? ??????? S
???????????????? 0.1, 0.2, 0.3, 0.4, 0.5,
0.6, 0.7, 0.8, 0.9, 1.0, 1.1, 1.2, 1.3,
1.4, 1.5 ??????? R ??????????????? 0-359
18
Case when Image Scale or Rotate
  • ????????????????? Scale ??? Rotation
    ??????????????????????????????????????????????????
    ???????? ??????

xc x rcos(?) yc y rsin(?)
xc x r S cos(? ?) yc y r S
sin(? ?)
19
Case when Image Scale or Rotate
  • Algorithm
  • 1. For each (?, r) from R-Table
  • For each S 0.3 to 1.5
  • For each ? 0 to 359 find
  • Arrayxc, yc, S, ?
  • Look for Maximum vote in 4-D Array

xc x r S cos(? ?) yc y r S
sin(? ?)
20
Example
21
Example
22
Example
23
(No Transcript)
24
Example
Write a Comment
User Comments (0)
About PowerShow.com