Content based watermarking of images

With the rapid growth of networked multimedia data systems, cop~-rightProtection of proprietary multimedia work has gained importance. Inw-ting a robust and invisible signal ( tLIafermar-k) that clearly identities the owner or the recipient is beginning to emerge as the solution. We present a novel invisible and robust. watermarking technique for images that can be e&ly extended for video data. Previous wat ermarking research have only partially used the results of the human visual system (Ht7S) studies done to evaluate the JPEG quantization table. This aiso does not provide a framework for the spatial domain water-marking methods. We propose a new w’ay of analyzing the noise sensitivity of wet-y pixel bzsed on the local region image content, such as tex%ure, edge and Iumimmce information. This results in a just rwticmhle distortion mzsk for the image to be watermarked. Then each bit of the watermark is .qmacl .~atially and shaped by a pseudo-noise sequence .mch that its amplitude is kept below the noise sensitivity of the pixel into which it is embedded. It. can be either embedded in the spatial domain or can be DCT coded to be embedded in the tramform domainExperimental results show that the resistant watermark is resistant to various attacks such as JF’EG compression. cropping, addition of noise and is perceptually invisible.