Watermarking Citra Digital Berwarna Pada Ruang Warna YIQ Menggunakan Contourlet Transform dan Discrete Cosine Transform Agus Setiawan (1022044) Email: Radiancharlot@hotmail.com Program Studi Teknik Elektro, Fakultas Teknik Jl. Prof. Drg. Suria Sumantri 65, Bandung 40164, Indonesia ABSTRAK Keamanan data dalam dunia digital sangat penting karena banyak pengguna internet yang melakukan pengambilan data digital. Untuk itu diperlukan suatu cara untuk menunjukkan keaslian data multimedia yang dikenal dengan istilah Digital Watermarking. Pada Tugas Akhir ini direalisasikan watermarking citra digital berwarna pada ruang warna YIQ menggunakan Contourlet Transform dan Discrete Cosine Transform. Citra host RGB, dikonversi ke ruang warna YIQ. Komponen Q dari citra host dilakukan Transformasi Contourlet level 2, lalu pilih subband bawah kiri kemudian dilakukan DCT. Lalu bangkitkan 6 bit pseudorandom sequence untuk bit 0 dan bit 1 untuk mewakili bit watermark. Selanjutnya bit watermark hasil Arnold Scrambling yang diwakili oleh pseudorandom sequence disisipkan dalam koefisien DCT midband. Watermark dapat diekstraksi kembali dengan mengkorelasikan koefisien DCT midband dengan masing-masing pseudorandom sequence. Hasil percobaan menunjukkan bahwa, citra ber-watermark memiliki rata-rata nilai PSNR > 30dB dan MOS > 4.3 (citra ber-watermark mirip dengan citra host). Watermark tahan terhadap kompresi JPEG dengan faktor kualitas Q = 8 hingga Q = 11, penambahan Gaussian noise sebesar 0.008, 0.009, 0.01, dan 0.02, Cropping dengan persentase 25%, 30%, dan 50% (bergantung pada daerah yang di crop), Salt & pepper noise dengan density 0.008, 0.009, 0.01, 0.02 dan sharpening. Watermark tidak tahan terhadap Rotation (90 0, 180 0, 270 0 ), Median filtering (3x3,5x5), dan Scaling 50%. Kata Kunci: Digital Watermarking, Countourlet Transform, DCT. i
Digital Color Image Watermarking In YIQ Colour Using Contourlet Transform and Discrete Cosine Transform Agus Setiawan (1022044) Email: Radiancharlot@hotmail.com Department of Electrical Engineering, Faculty of Engineering Maranatha Christian University Jl. Prof. Drg. Suria Sumantri 65, Bandung 40164, Indonesia ABSTRACT Security of data in the digital world is very important because many Internet users who perform digital data retrieval. For that we need a way to show the authenticity of multimedia data which is known as Digital Watermarking. In this final project, it is implemented digital color image watermarking using Contourlet Transform and Discrete Cosine Transform. The host image readed, then convert RGB to YIQ colour space. Select Q component then transformed by contourlet level 2, select the left bottom subband, transfromed again with DCT. Then generate 6-bits pseudorandom sequence for bit 0 and bit 1 to represent a watermark bit, then insert the Arnold scrambled watermark bit into Midband DCT coefficients. Watermark can be extracted by correlating Midband DCT coefficients with 6 bits pseudorandom sequence. Experimental results show that watermarked images have average PSNR value greater than 30dB and MOS value greater than 4.3 (watermarked image quite similar to host image). Watermark is robust to common image processing, such as JPEG compression 8 up to 11 quality factor, adding Gaussian noise 0.008, 0.009, 0.01, and 0.02, Cropping 25%, 30%, and 50% (depent on cropping position of image), adding Salt & pepper noise 0.008, 0.009, 0.01, 0.02 density, and sharpening. Watermark can t withstand to image processing such as Rotation (90 0, 180 0, 270 0 ), median filtering (3x3,5x5) and scaling 50%. Keywords: Digital Watermarking, Contourlet Transform, DCT. ii
DAFTAR ISI ABSTRAK... i ABSTRACT... ii KATA PENGANTAR... iii DAFTAR ISI...v DAFTAR GAMBAR... vii DAFTAR TABEL... viii BAB I PENDAHULUAN...1 1.1 Latar Belakang...1 1.2 Perumusan Masalah...2 1.3 Tujuan...2 1.4 Pembatasan Masalah...2 1.5 Sistematika Penulisan...3 BAB II LANDASAN TEORI...4 2.1 Teori Citra...4 Citra Digital...4 Klasifikasi Citra Digital...5 Ruang Warna RGB (Red Green Blue)...7 2.1.4 Ruang Warna YIQ......7 2.2 Digital Watermarking...8 2.3 Discrete Cosine Transform (DCT) Dimensi Dua (2-D DCT)...8 2.4 Zig-zag Scanning...10 2.6 Contourlet Transform...10 2.7 Kualitas Citra...13 Peak Signal to Noise Rasio (PSNR)...13 Mean Opinion Score (MOS)...14 Normalized Cross Corelation (NCC)...15 2.7.4 Bit Correlation Ratio (BCR)...15 BAB III PERANCANGAN PERANGKAT LUNAK...17 v
3.1 Proses Penyisipan Watermark...17 Diagram Blok Proses Penyisipan Watermark...17 Diagram Alir Proses Penyisipan Watermark...20 3.2 Proses Ekstraksi Watermark...21 Diagram Blok Proses Ekstraksi Watermark...22 Diagram Alir Proses Ekstraksi Watermark...23 3.3 Rancangan Tampilan GUI (Graphic User Interface) Program...25 BAB IV DATA PENGAMATAN DAN ANALISIS...26 4.1 Tampilan Hasil Rancangan GUI (Graphic User Interface) Program...26 4.2 Prosedur Pengujian...27 Pengukuran Kualitas Citra...30 4.3 Pengujian Ketahanan Watermark...31 Gaussian Noise...32 Rotation...34 Cropping...35 JPEG Compression...37 Scaling...39 Salt and Pepper noise...41 Median Filtering...43 Sharpening...44 BAB V SIMPULAN DAN SARAN...45 5.1 Simpulan...45 5.2 Saran...46 DAFTAR PUSTAKA...47 LAMPIRAN A... A-1 A.1 Program Utama Watermarking... A-1 A.2 Program Tambahan... A-8 A.2.1 Zigzag Scanning... A-8 A.2.2 Inverse Zigzag Scanning... A-9 A.2.3 Arnold Transform... A-10 LAMPIRAN B...B-1 vi
DAFTAR GAMBAR Gambar 2.1 Citra Biner... 5 Gambar 2.2 Citra Grayscale... 6 Gambar 2.3 Citra Berwarna... 6 Gambar 2.4 Ruang Warna RGB... 7 Gambar 2.5 Proses Zig-zag Scanning... 9 Gambar 2.6 Contourlet Filter Bank...11 Gambar 2.7 Level Dekomposisi Contourlet...12 Gambar 3.1 Diagram Blok Proses Penyisipan... 16 Gambar 3.2 Dekomposisi Contourlet level 0... 17 Gambar 3.3 Diagram Alir Proses Penyisipan... 19 Gambar 3.4 Diagram Blok Proses Ekstraksi Watermark... 21 Gambar 3.5 Diagram Alir Proses Ekstraksi Watermark... 22 Gambar 3.6 Rancangan Tampilan GUI... 24 Gambar 4.1 Tampilan Hasil Rancangan GUI Program... 25 Gambar 4.2 Grafik PSNR dengan K= 10 hingga K = 35... 28 Gambar 4.3 Grafik NCC dengan K= 10 hingga K = 35... 28 vii
DAFTAR TABEL Tabel 2.1 Kriteria Mean Opinion Score (MOS)... 14 Tabel 4.1 Hasil Penyisipan dan Ekstraksi Watermark... 26 Tabel 4.2 Penilaian MOS Citra Airplane, Pepper, dan Boats (α = 10, 20, 30, 40, dan 50)... 28 Tabel 4.3 Watermark Hasil Ekstraksi Setelah Dilakukan Proses Penambahan Gaussian Noise... 33 Tabel 4.4 Watermark Hasil Ekstraksi Setelah Dilakukan Proses Rotation... 35 Tabel 4.5 Watermark Hasil Ekstraksi Setelah Dilakukan Proses Scaling... 37 Tabel 4.6 Watermark Hasil Ekstraksi Setelah Dilakukan Proses JPEG Compression... 39 Tabel 4.7 Watermark Hasil Ekstraksi Setelah Dilakukan Proses Cropping... 41 viii