Fuzzy for Image Processing

1
Fuzzy for Image Processing

• Penyusun
• Tri Nurwati
• (Dari berbagai sumber)

2
Fuzzy Image Processing

• Fuzzy image processing is the collection of all
  approaches that understand, represent and process
  the images, their segments and features as fuzzy
  sets. The representation and processing depend on
  the selected fuzzy technique and on the problem
  to be solved.(From Tizhoosh, Fuzzy Image
  Processing, Springer, 1997)

3
Struktur pengolahan citra dengan fuzzy
4
Proses pembuatan fuzzy pada pengolahan citra

• Tidak seperti penggunakan logika fuzzy di suatu
  plant, untuk pengolahan citra pembuatan fuzzy
  melalui proses
• coding of image data (fuzzification)
• the middle step (modification of membership
  values
• decoding of the results (defuzzification)

5
Proses pembuatan fuzzy pada pengolahan citra

• Setelah data citra ditransformasikan dari level
  gray ke dalam membership function
  (fuzzification), dalam proses ini dibutuhkan
  ketelitian dalam pengelompokan dan penentuan
  nilai membership input dan output

6
(No Transcript)
7
Kelebihan pengolahan citra dengan menggunakan
logika fuzzy

• Teknik logika fuzzy sangat mumpuni dalam
  pemrosesan/pengolahan dan representatif
  pengetahuan (rule)
• Teknik logika Fuzzy dapat mengatur keambiguan
  (mirip) dan hal-hal yang relatif

8
Kelebihan pengolahan citra dengan menggunakan
logika fuzzy

• Teori set fuzzy mempunyai kelebihan dapat
  mempresentasikan dan memproses pengetahuan
  pengguna dalam bentuk aturan it-then

9
(No Transcript)
10
Contohcolour yellow, orange, red, violet,
blue
11
Contohwarna gray gelap, gray, dan terang
12
Aplikasi

• Histogram-based gray-level fuzzification (or
  briefly histogram fuzzification)contoh
  Perbaikan ketajaman warna image (seperti gambar
  panda di atas)
• Local fuzzification (contoh deteksi tepi)
• Feature fuzzification (Scene analysis, object
  recognition)

13
Perbaikan Image dengan Fuzzy

• many researchers have applied the fuzzy set
  theory to develop new techniques for contrast
  improvement

14
Langkah-langkah

• 1.1. Contrast Improvement with INT- Operator
• Langkah
• a.menentukan fungsi membership
• b.Mengubah nilai membership
• c.Membuat skala warna gray

15
1.2. Contrast Improvement using Fuzzy Expected
Value (Craig and Schneider 1992)

• 1. Step Calculate the image histogram
• 2. Step Determine the fuzzy expected value (FEV)
• 3. Step Calculate the distance of gray-levels
  from FEV
• 4. Step Generate new gray-levels

16
1.3. Contrast Improvement with Fuzzy Histogram
Hyperbolization (Tizhoosh 1995/1997)

• 1. Step Setting the shape of membership function
  (regrading to the actual image)
• 2. Step Setting the value of fuzzifier Beta (a
  linguistic hedge)
• 3. Step Calculation of membership values
• 4. Step Modification of the membership values by
  linguistic hedge
• 5. Step Generation of new gray-levels

17
1.4. Contrast Improvement based on Fuzzy If-Then
Ruels (Tizhoosh 1997)

• Step Setting the parameter of inference system
  (input features, membership functions,..)
• Step Fuzzification of the actual pixel
  (memberships to the dark, gray and bright sets of
  pixels)
• .

18
1.4. Contrast Improvement based on Fuzzy If-Then
Ruels (Tizhoosh 1997)

• 3. Step Inference (e.g. if dark then darker, if
  gray then gray, if bright then brighter)
• 4. Step Defuzzification of the inference result
  by the use of three singletons

19
1.5. Locally Adaptive Contrast Enhancement
(Tizhoosh et al. 1997)

• In many cases, the global fuzzy techniques fail
  to deliver satisfactory results. Therefore, a
  locally adaptive implementation is necessary to
  achieve better results. See some examples and a
  comparison with calssical approach.

20
(No Transcript)
21
(No Transcript)
22
(No Transcript)
23
(No Transcript)
24
Deteksi Tepi

• Perbaiki dengan rumus di bawah

25
Deteksi Tepi
26
Contoh Hasil Deteksi Tepi
27
Segmentasi Image dengan Fuzzy
28
Segmentasi Image dengan Fuzzy
29
(No Transcript)
30
Contoh segmentasi