Kompresi Citra Lossless Menggunakan Transformasi Bilangan Bulat Dan Differential Encoding

Hendra, - (2017) Kompresi Citra Lossless Menggunakan Transformasi Bilangan Bulat Dan Differential Encoding. Doctoral thesis, Institut Teknologi Sepuluh Nopember.

[thumbnail of 5113301001-Disertation.pdf] Text
5113301001-Disertation.pdf - Published Version
Restricted to Repository staff only

Download (16MB) | Request a copy


citra lossless adalah Predictive Coding, Burrows-Wheeler Transform, dan Variable Block Size Segmentation. Dalam algoritmanya, metode-metode kompresi lossless menggunakan entropy coding untuk mengkodekan citra dengan kode yang lebih efisien. Penelitian ini mengembangkan metode kompresi citra lossless dengan pendekatan yang berbeda yaitu menggunakan konsep reversible watermarking yang dibangun dari transformasi bilangan bulat Reversible Contrast Mapping (RCM) dan Reversible Low Contrast Mapping (RLCM). Penelitian memperkenalkan tiga skema kompresi citra lossless menggunakan transformasi bilangan bulat yaitu skema reversible watermarking dan pendekatan coding dalam skema Average Encoding (AE) dan Hierarchy Average Encoding (HAE). Pada pendekatan coding, penelitian ini mengkolaborasikan RLCM dan Differential Encoding. Metode Cyclic Reversible Low Contrast Mapping (CRLCM) merupakan metode dengan rasio kompresi tertinggi dalam skema reversible watermarking. Pada skema AE, penelitian ini memperkenalkan empat metode coding yaitu 1RLCM (one step RLCM)-Up, Ext-1RLCM-Up, 1RLCM-Dw, dan 1RLCM-0. Metode yang memiliki rasio kompresi tertinggi pada skema ini adalah 1RLCM-0. Skema HAE diimplementasikan dalam dua metode yaitu metode dasar dan MAPHAE. Kompleksitas Algoritma metode-metode dengan pendekatan coding sama dengan Golomb coding yaitu ܱ(݊). Rasio kompresi metode 1RLCM-0 pada skema AE dan skema HAE umumnya sama dengan metode kompresi Golomb coding untuk citra standar tetapi metode-metode ini memiliki rasio kompresi yang lebih besar untuk citra medik. Implementasi 1RLCM-0 pada metode kompresi citra JPEG-LS dengan mengganti metode adaptive-Golomb yang digunakan pada metode tersebut menghasilkan rasio kompresi sebanding dengan metode JPEG-LS original
Lossless image compression methods were developed to save and transmit images without losing any detailed information of the images. Applications of these compression methods are for compressing the satellite images, aerial photographs, and medical data. The most common approaches used in this compression type are predictive coding, Burrows-Wheeler Transform, and variable block size segmentation. In the algorithms, lossless image compression methods use entropy coding for more efficient compression result. This research develops a new lossless image compression method using a different approach that exploiting the concept of the reversible watermarking method for the integer transform of Reversible Contrast Mapping (RCM) and Reversible Low Contrast Mapping (RLCM). This research introduces three lossless image compression schemes, i.e. reversible watermarking scheme, coding approach by Average Encoding (AE), and Hierarchy Average Encoding (HAE) Scheme. The coding approach collaborates the RLCM transform and the Differential Encoding. On the reversible watermarking scheme, the Cyclic Reversible Low Contrast Mapping (CRLCM) has the highest compression ratio. In AE scheme, this research introduces four coding methods, i.e. 1RLCM (one step RLCM)-Up, Ext-1RLCMUp, 1RLCM-Dw, and 1RLCM-0. The 1RLCM-0 has the highest compression ratio for this scheme. The HAE scheme is implemented using two approaches, i.e. the basic method and MAP-HAE. The algorithm complexity of the coding approaches is equal to Golomb coding complexity, which is ܱ(݊). The compression ratio of the 1RLCM-0 for AE scheme and HAE scheme is generally equal to Golomb coding’s for the standard images, but the proposed methods have significantly higher compression ratio for medical images. Implementation of the 1RLCM-0 in JPEG-LS compression method by replacing the adaptive-Golomb coding in the method results in similar compression ratio compared to the JPEG-LS original

Item Type: Thesis (Doctoral)
Additional Information: RDIf 621.367 Hen k
Uncontrolled Keywords: Kompresi citra lossless, Transformasi bilangan bulat, Differential Encoding, Reversible Contrast Mapping, Reversible Low Contrast Mapping, Average Encoding, Hierarchy Average Encoding
Subjects: Q Science > QA Mathematics
Q Science > QA Mathematics > QA76.87 Neural networks (Computer Science)
T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK5105.546 Computer algorithms
Divisions: Faculty of Information Technology > Informatics Engineering > (S3) PhD Thesis
Depositing User: Mr. Tondo Indra Nyata
Date Deposited: 31 May 2017 07:21
Last Modified: 21 Jun 2023 02:01
URI: http://repository.its.ac.id/id/eprint/41450

Actions (login required)

View Item View Item