UPT. PERPUSTAKAAN
Institut Teknologi Sepuluh Nopember Surabaya
Kampus ITS Sukolilo - Surabaya 60111
Phone
:
031-5921733 , 5923623
Fax
:
031-5937774
E-mail
:
libits@its.ac.id
Website
:
http://library.its.ac.id
Support (Customer Service) :
timit_perpus@its.ac.id
Welcome..guys!
Have a problem with your access?
Please, contact our technical support below:
LIVE SUPPORT
Davi Wahyuni
Tondo Indra Nyata
Anis Wulandari
Ansi Aflacha
ITS » Master Theses » Teknik Informatika Posted by tondoindra@gmail.com at 21/04/2016 14:46:23 • 2352 Views
SEGMENTATION OF TUNA FISH IMAGE BY MAHALANOBIS HISTOGRAM THRESHOLDING AND MAHALANOBIS FUZZY C-MEANS
Author : KASWAR, ANDI BASO ( 5114201010 )
ABSTRAK
segmentasi. Secara visual citra ikan tuna memiliki varian warna yang tinggi mulai
dari bagian biru gelap hingga biru terang atau putih. Jika citra ikan tuna dipetakan
ke dalam ruang fitur maka piksel yang menyusun area tubuh ikan tuna akan
membentuk kelompok piksel hyperellipsoid. Algoritma segmentasi Fuzzy C-Means
berbasis jarak Mahalanobis dapat digunakan untuk mensegmentasi citra yang
memiliki karakteristik tersebut. Namun inisialisasi derajat keanggotaan dan
centroid klaster secara random mengakibatkan proses pengklasteran piksel menjadi
tidak efisien dalam hal iterasi dan waktu komputasi.
Penelitian ini mengusulkan metode baru untuk segmentasi citra ikan tuna
dengan Mahalanobis Histogram Thresholding M-HT dan Mahalanobis Fuzzy CMeans
MFCM. Metode yang diusulkan terdiri atas tiga tahapan utama yaitu
Inisialisasi centroid untuk mendapatkan centroid awal tiap klaster pengklasteran
piksel untuk mengelompokkan piksel background dan objek dan peningkatan
akurasi untuk meningkatkan akurasi hasil pengklasteran piksel.
Berdasarkan hasil uji coba diperoleh rata-rata jumlah iterasi sebanyak 66
iterasi dengan waktu segmentasi rata-rata 16203 detik. Rata-rata Akurasi sebesar
9854 dengan tingkat Missclassification Error sebesar 146. Berdasarkan hasil
yang diperoleh maka dapat disimpulkan bahwa metode yang diusulkan dapat
meningkatkan efisiensi dalam hal jumlah iterasi dan waktu segmentasi. Selain itu
metode yang diusulkan dapat memberikan hasil segmentasi yang lebih akurat
dibandingkan dengan metode konvensional.
ABSTRACT
One of important phases in classification system of tuna fish is
segmentation phase. Visually image of tuna fish have high color variant from dark
blue until light blue or white. If image of Tuna fish are mapped in feature space
pixel that arrange the body of tuna fish will form group of pixel of hyperellipsoid.
Fuzzy C-Means segmentation algorithm based on Mahalanobis distance can be
used to segment image that has the characteristic. However initialization of clusters
centroid randomly causes segmentation process to be inefficient in terms of
iterations and computational time.
In this study we proposed a new method for segmentation of tuna image
with Mahalanobis Histogram Thresholding M-HT and Mahalanobis Fuzzy Cmeans
MFCM. The proposed method consist of three important phases namely
Initialization centroid to obtain centroid of each cluster pixel clustering to group
the background pixels and object pixels and improvement of accuracy to improve
the accuracy of pixels clustering results.
Based on the experiment it is obtained average amount of iteration as
many as 66 iteration with time of segmentation average as long as 162.03 second.
While average Accuracy as big as 98.54 with level of Missclassification Error as
big as 1.46. Based on the obtained result it can be concluded that the proposed
method can improve the efficiency in terms of the number of iterations and time
segmentation. Besides that the proposed method can give accurate segmentation
result compared with the conventional method.
Keywords:
Citra Ikan Tuna, Segmentasi, Fuzzy Clustering, Histogram
Thresholding, Jarak Mahalanobis.
Subject
: Segmentasi Gambar
Contributor
Dr. Agus Zainal Arifin, S.Kom., M.Kom.
Arya Yudhi Wijaya, S.Kom., M.Kom
Date Create
: 21/04/2016
Type
: Text
Format
: PDF
Language
: Indonesian
Identifier
: ITS-Master-51103150001614
Collection ID
: 51103150001614
Call Number
: RTif 006.42 Kas s
Source Master Theses Of Informatics Engineering RTif 006.42 Kas s, 2016
Coverage ITS Community
Rights Copyright @2016 by ITS Library. This publication is protected by copyright and per obtained from the ITS Library prior to any prohibited reproduction, storage in a re transmission in any form or by any means, electronic, mechanical, photocopying, reco For information regarding permission(s), write to ITS Library