Strategi Crossover Pada Algoritma Differential Evolution Berdasarkan Similaritas Antar-Cluster Graylevel Untuk Automatic Multilevel Image Thresholding

Umam, Khoirul (2015) Strategi Crossover Pada Algoritma Differential Evolution Berdasarkan Similaritas Antar-Cluster Graylevel Untuk Automatic Multilevel Image Thresholding. Masters thesis, Institut Teknology Sepuluh Nopember.

[thumbnail of 5113201017-Master Thesis.pdf] Text
5113201017-Master Thesis.pdf - Published Version
Restricted to Repository staff only

Download (10MB) | Request a copy

Abstract

Pendekatan automatic multilevel image thresholding memiliki area pencarian
solusi optimal yang luas karena dapat menentukan jumlah dan posisi
threshold secara bersamaan. Pencarian solusi optimal kasus automatic multilevel
image thresholding menggunakan strategi standar algoritma Differential Evolution
(DE) dapat mengalami penurunan efisiensi karena kemampuan konvergensinya
yang melambat. Oleh karena itu dibutuhkan strategi yang dapat membatasi
area pencarian solusi optimal agar proses optimasi menjadi efisien. Pada penelitian
ini diusulkan strategi baru untuk operasi crossover algoritma DE berdasarkan
similaritas antar-cluster graylevel untuk kasus automatic multilevel image thresholding.
Strategi tersebut membatasi area pencarian dengan cara hanya merekombinasi
cluster-cluster graylevel dengan tingkat similaritas yang kecil. Perhitungan
similaritas antar-cluster graylevel dilakukan dengan mengintegrasikan nilai interclass
dan intra-class variance dari cluster graylevel yang saling bertetangga. Uji
coba dilakukan pada data citra abu-abu dari Berkeley Segmentation Dataset
(BSDS500). Hasil pengujian menunjukkan bahwa DE yang menggunakan strategi
crossover usulan memberikan hasil segmentasi dengan rata-rata misclassification
error 40,47% dan hanya membutuhkan rata-rata 1106 generasi untuk menemukan
solusi optimal. Hasil tersebut lebih baik dibandingkan dengan strategi crossover
yang tidak memperhitungkan similaritas antar-cluster graylevel.
=======================================================================================================
Automatic multilevel image thresholding approach has wide optimal solution
search space due to its ability to determine thresholds number and positions,
simultaneously. Searching its optimal solution using standard Differential
Evolution (DE) algorithm can decrease its efficiency due to slow convergence.
Therefore, a strategy that can restrict the search space is needed in order for optimizations
being efficient. In this paper we propose a novel strategy of DE's crossover
operator based on graylevel clusters similarity for automatic multilevel
image thresholding. We restrict the search space by only recombining graylevel
clusters which have small similarity. Graylevel clusters similarity is performed by
computing the inter-class and intra-class variance of adjacent graylevel clusters.
Experiments on grayscale image of Berkeley Segmentation Dataset (BSDS500)
show that the proposed crossover strategy can generate segmented images with
misclassification error of 40.47% and only requires the average of 1106 generations
to find the optimal solution. It is better than crossover strategies that not
compute the graylevel clusters similarity.

Item Type: Thesis (Masters)
Additional Information: RTIf 621.367 Uma s
Uncontrolled Keywords: automatic multilevel image thresholding, crossover, Differential Evolution, similaritas antar-cluster
Subjects: T Technology > TA Engineering (General). Civil engineering (General) > TA1637 Image processing--Digital techniques
Divisions: Faculty of Information Technology > Informatics Engineering > 55101-(S2) Master Thesis
Depositing User: Mr. Tondo Indra Nyata
Date Deposited: 05 Nov 2019 02:06
Last Modified: 05 Nov 2019 02:06
URI: http://repository.its.ac.id/id/eprint/71595

Actions (login required)

View Item View Item