Upgrade to Pro — share decks privately, control downloads, hide ads and more …

Cellular Edge Detection

Kath
October 25, 2017
0
40

Cellular Edge Detection

Kath

October 25, 2017
Tweet

More Decks by Kath

See All by Kath
Inn Lab
kathsaz
0
51
Click & Collect
kathsaz
0
35
Foodie.me
kathsaz
0
33
Innovation Lab
kathsaz
0
32
Innovation Lab
kathsaz
0
52
Design Thinking
kathsaz
0
46
Innovation Lab
kathsaz
0
92
Smart Campus - Re-imagining the user experience
kathsaz
0
50
Big Data
kathsaz
0
43

Featured

See All Featured
Fireside Chat
paigeccino
22
2.7k
The Psychology of Web Performance [Beyond Tellerrand 2023]
tammyeverts
15
1.6k
GraphQLとの向き合い方2022年版
quramy
33
13k
CoffeeScript is Beautiful & I Never Want to Write Plain JavaScript Again
sstephenson
155
14k
How GitHub Uses GitHub to Build GitHub
holman
468
290k
Building Applications with DynamoDB
mza
88
5.7k
We Have a Design System, Now What?
morganepeng
43
6.8k
Teambox: Starting and Learning
jrom
128
8.4k
Bootstrapping a Software Product
garrettdimon
PRO
302
110k
Design by the Numbers
sachag
274
18k
Designing the Hi-DPI Web
ddemaree
276
33k
個人開発の失敗を避けるイケてる考え方 / tips for indie hackers
panda_program
67
14k

Transcript

  1. CELLULAR EDGE DETECTION FOR GRAYSCALE & BINARY IMAGES

  2. None
  3. None

  4. CELLULAR AUTOMATON BINARY IMAGE

  5. CELLULAR AUTOMATON GREYSCALE IMAGE

  6. LEARNING AUTOMATON Von-neumann Moore

  7. 50% 50% LEARNING AUTOMATON Von-neumann Moore

  8. 10% 90% LEARNING AUTOMATON Von-neumann Moore

  9. None

  10. 1200x800 Original 1200x800 Output image Cellular automata 1200x800 Learning automata

    1200x800

  11. 1200x800 Output image Cellular automata 1200x800 Learning automata 1200x800 ..update

    output image every generation Execute rules for N generations..

  12. Final 1200x800 1200x800 Output image

  13. EXPERIMENTS

  14. None
  15. None

  16. CED CANNY PREWITT

  17. None

  18. CED CANNY PREWITT

  19. CED CANNY PREWITT MSE PSNR (dB) MSE PSNR (dB) MSE

    PSNR (dB) 1.368 6.8011 1.369 6.8075 1.367 6. 8017 3.493 6.7320 3.491 6.7346 3.488 6.7382 1.804 5.6016 1.802 5.6056 1.800 5.6120

  20. LIMITATIONS

  21. Manual adjustment of threshold Defeats the purpose (maybe) Slow for

    big images
  22. None

  23. SUPPORTING SLIDES

  24. Pseudocode

  25. Reinforcement signal