., Munawaroh (2014) Implementasi Segmentasi Citra Dengan Metode Fuzzy Co-Clustering Dan Particle Swarm Optimization Pada Ruang Warna Cielab. Undergraduate thesis, Institut Teknologi Sepuluh Nopember Surabaya.
Preview |
Text
5110100160-Undergraduate_Thesis.pdf - Published Version Download (3MB) | Preview |
Abstract
Segmentasi citra merupakan suatu metode partisi terhadap citra menjadi beberapa bagian yang homogen berdasarkan kemiripan tertentu. Proses segmentasi sangat penting karena hasil segmentasi mempengaruhi hasil dari proses yang akan dilakukan selanjutnya seperti pengenalan pola. Dalam melakukan segmentasi, telah banyak metode yang digunakan diantaranya adalah segmentasi dengan histogram, segmentasi dengan deteksi tepi, segmentasi dengan Fuzzy C-Means (FCM) dan masih banyak lagi metode lainnya.
Pada tugas akhir ini, metode yang akan diimplentasikan dalam proses segmentasi citra berwarna adalah algoritma Fuzzy Co-Clustering dengan dual fuzzy yaitu dengan dua fungsi keanggotaan yaitu objek dan fitur. Algoritma Fuzzy Co-Clustering For Images (FCCI) dikembangkan dengan menggabungkan jarak antara setiap fitur titik data dengan fitur cluster center sebagai ukuran ketidakmiripan (dissimilarity) dan entropy dari objek dan fitur sebagai kondisi regularisasi dalam fungsi objektif. Untuk mengoptimalkan threshold dalam proses segementasi citra ini digunakan juga algoritma Particle Swarm Optimization dengan modifikasi perilaku bakteri. Dengan adanya algoritma ini hasil segmentasi diharapkan lebih optimal
===================================================================================================
Image segmentation is a method of the digital image processing to divide an image into some homogeneous regions based on similarity criteria. Segmentation process is very important due to its result gives a significant effect for the next step such as pattern recognition. Many methods have been implemented for segmentation process such as segmentation based on histogram, segmentation using edge detection, segmentation using Fuzzy C-Means (FCM) and so forth.
On this final project,the method which is going to be implemented for image segmentation process is Fuzzy Co-Clustering algorithm with dual fuzzy membership functions. Fuzzy Co-Clustering Algorithm For Images (FCCI) is developed by using combination of distance for every data point features and feature cluster center as a measure of dissimilarity.For condition regularization of objective function, we use entropy of object and feature.Beside that, we also use Particle Swarm Optimization with modification of bacterial behavior for threshold optimization. By using this heuristic algorithm, the result of segmentation is expected to be more optimal
| Item Type: | Thesis (Undergraduate) |
|---|---|
| Additional Information: | RSIf 621.367 Mun i |
| Uncontrolled Keywords: | Citra; Fuzzy Co-Clustering; Particle Swarm; Optimization ; Cielab |
| Subjects: | T Technology > TA Engineering (General). Civil engineering (General) > TA1637 Image processing--Digital techniques |
| Divisions: | Faculty of Information Technology > Informatics Engineering > 55201-(S1) Undergraduate Thesis |
| Depositing User: | EKO BUDI RAHARJO |
| Date Deposited: | 08 Jan 2021 04:08 |
| Last Modified: | 08 Jan 2021 04:08 |
| URI: | http://repository.its.ac.id/id/eprint/82387 |
Actions (login required)
 |
View Item |