Compressive Sensing Berbasis Dct-dwt Untuk Kompresi Watermark Pada Watermarking Citra Digital Dengan Domain Swt-svd

Abstract

Abstrak Image watermarking adalah proses perlindungan hak cipta dari suatu objek (gambar) dengan memberikan tanda di dalam objek tersebut. Teknik image watermarking, dikatakan baik apabila tanda yang disisipkan tidak tampak oleh kasat mata dan objek pembawanya tidak mengalami penurunan kualitas serta tanda yang disisipkan harus tahan terhadap berbagai pengolahan sinyal. Pada tugas akhir ini di implementasikan sistem image watermarking pada watermark dan host dalam bentuk citra hitam putih dan Red Green Blue (RGB). Pada sistem image watermarking menggunakan metode Compressive Sensing (CS) berbasis Discrete Cosine Transform-Discrete Wavelet Transform (DCT-DWT), Stationary Wavelet Transform-Singular Value Decomposition (SWT-SVD) dan proses rekontruksi menggunakan algoritme Orthogonal Matching Pursuit (OMP). Pengujian dilakukan terhadap sistem yang di bangun dan telah dilakukan beberapa skenario pengujian dalam memperoleh analisis terkait dengan performansi sistem yang mampu dihasilkan. Pengaruh CS pada pengujian menghasilkan nilai Mean Square Error(MSE) 4,396208 , Peak Signal to Noise Ratio (PSNR) 41,70416 dan Bit Error Rate (BER) 0,185565. Dan pengaruh tanpa CS pada pengujian menghasilkan nilai Mean SquareError(MSE) 3,53102, Peak Signal to Noise Ratio (PSNR) 42,6518 dan Bit Error Rate (BER) 0,19686 Kata kunci: Image watermarking, CS, DCT-DWT, SWT-SVD, OMP, MSE, PSNR, BER Abstract Image watermarking is the process of protecting the copyright of an object (image) by giving a sign inside the object. Image watermarking technique is said to be good if the inserted sign does not appear to be visible and the carrying object does not experience a decrease in quality and the inserted sign must be resistant to various signal processing. This final project implements an image watermarking system on watermarks and hosts in the form of black and white and Red Green Blue (RGB) images. The image watermarking system uses the Compressive Sensing (CS) method based on Discrete Cosine Transform-Discrete Wavelet Transform (DCT-DWT), Stationary Wavelet Transform-Singular Value Decomposition (SWT-SVD) and the reconstruction process using the Orthogonal Matching Pursuit (OMP) algorithm. Tests are carried out on systems that are built and several test scenarios have been carried out in ISSN : 2355-9365 e-Proceeding of Engineering : Vol.6, No.1 April 2019 | Page 492 obtaining analysis related to the system performance that can be produced. The effect of CS on the test results in the value of Mean Square Error (MSE) 4,396208, Peak Signal to Noise Ratio (PSNR) 41,70416 and Bit Error Rate (BER) 0,185565. And the effect without CS on the test resulted in the value of MSE 3.53102, PSNR 42.6518 and BER 0.19686 Keywords:Image watermarking, CS, DCT-DWT, SWT-SVD, OMP, MSE, PSNR, BER