30
Fuzzy for Image Processing Penyusun: Tri Nurwati (Dari berbagai sumber)

Fuzzy for Image Processing

  • Upload
    rusty

  • View
    60

  • Download
    14

Embed Size (px)

DESCRIPTION

fuzzy logic. Fuzzy for Image Processing. Penyusun: Tri Nurwati (Dari berbagai sumber). fuzzy logic. Fuzzy Image Processing . - PowerPoint PPT Presentation

Citation preview

Page 1: Fuzzy for Image Processing

Fuzzy for Image ProcessingPenyusun:Tri Nurwati

(Dari berbagai sumber)

Page 2: Fuzzy for Image Processing

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)

Page 3: Fuzzy for Image Processing

Struktur pengolahan citra dengan fuzzy

Page 4: Fuzzy for Image Processing

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)

Page 5: Fuzzy for Image Processing

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

Page 6: Fuzzy for Image Processing
Page 7: Fuzzy for Image Processing

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

Page 8: Fuzzy for Image Processing

Kelebihan pengolahan citra dengan menggunakan logika

fuzzy• Teori set fuzzy mempunyai

kelebihan dapat mempresentasikan dan memproses pengetahuan pengguna dalam bentuk aturan “it-then”

Page 9: Fuzzy for Image Processing
Page 10: Fuzzy for Image Processing

Contoh:colour = {yellow, orange, red, violet, blue}

Page 11: Fuzzy for Image Processing

Contoh:warna gray: gelap, gray, dan terang

Page 12: Fuzzy for Image Processing

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)

Page 13: Fuzzy for Image Processing

Perbaikan Image dengan Fuzzy

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

Page 14: Fuzzy for Image Processing

Langkah-langkah1.1. Contrast Improvement with INT-

OperatorLangkah: a.menentukan fungsi membership

b.Mengubah nilai membership

c.Membuat skala warna gray

Page 15: Fuzzy for Image Processing

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

Page 16: Fuzzy for Image Processing

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

Page 17: Fuzzy for Image Processing

1.4. Contrast Improvement based on Fuzzy If-Then Ruels

(Tizhoosh 1997) 1. Step: Setting the parameter of inference system

(input features, membership functions,..)2. Step: Fuzzification of the actual pixel

(memberships to the dark, gray and bright sets of pixels)

.

Page 18: Fuzzy for Image Processing

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

Page 19: Fuzzy for Image Processing

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.

Page 20: Fuzzy for Image Processing
Page 21: Fuzzy for Image Processing
Page 22: Fuzzy for Image Processing
Page 23: Fuzzy for Image Processing
Page 24: Fuzzy for Image Processing

Deteksi Tepi• Perbaiki dengan rumus di bawah

Page 25: Fuzzy for Image Processing

Deteksi Tepi

Page 26: Fuzzy for Image Processing

Contoh Hasil Deteksi Tepi

Page 27: Fuzzy for Image Processing

Segmentasi Image dengan Fuzzy

Page 28: Fuzzy for Image Processing

Segmentasi Image dengan Fuzzy

Page 29: Fuzzy for Image Processing
Page 30: Fuzzy for Image Processing

Contoh segmentasi