A robust image watermarking using hybrid DCT and SLT

Abstract

Slantlet Transform (SLT) is an improved performance from Discrete Wavelet Transform (DWT). Based on characteristic on SLT; piecewise linear, two zero moments, and power quality for data compression. Those characteristic proves efficiency of SLT. One of the characteristic of SLT named power quality is better than Discrete Cosine Transform (DCT). DCT has a good signal compression. Their capability for embedding and resulting a robust image watermarking. In this paper, we proposed a hybrid technique in SLT-DCT. Our experimental result achieved high PSNR that proved good imperceptibility and robustness both in Peak Signal to Noise Ratio (PSNR) and Normalized Correlation (NC) that was tested in some image compression attack. This experiment has been done and yielded higher PSNR than previous research as compared in our results discussion. The experiments were performed using 5 pieces of gray scale image size of 512×512 jpg. The fifth image generating PSNR is greater than 40 dB. Barbara.jpg results the highest PNSR from of all the images that is 63.1920 dB and NC to attack JPEG compression is 1. This data proved that the proposed method produces robust image watermarking.