作者: H. Sadreazami , M. Omair Ahmad , M. N. S. Swamy
DOI: 10.1109/TCSII.2015.2468995
关键词: Gaussian noise 、 Pattern recognition 、 Mathematics 、 Artificial intelligence 、 Digital watermarking 、 Noise 、 Median filter 、 Detector 、 Salt-and-pepper noise 、 Statistical model 、 Computer vision 、 Watermark
摘要: In recent years, digital watermarking has facilitated the protection of copyright information through embedding hidden into content. this brief, for first time, a blind multichannel multiplicative color image scheme in sparse domain is proposed. order to take account cross correlation between coefficients bands domain, statistical model based on multivariate Cauchy distribution used. The then used derive an efficient closed-form decision rule watermark detector. Experimental results and theoretical analysis are presented validate proposed performance detector compared with that other detectors. demonstrate improved detection rate high robustness against commonly attacks such as JPEG compression, salt pepper noise, median filtering, Gaussian noise.