Watermarking Algorithm For Image Authentication And Recovery

With the rapid development of information technology, digital products can now be produced, copied, accessed and distributed much easier, which also brings many problems, such as piracy, unauthorized access, copying, tampering and dissemination infringement. Copyright protection of digital products and the authenticity and integrity of the certification is becoming an urgent problem, at this moment, digital watermarking technology emerged. Digital watermarking technology embeds some special information into the multimedia content through a certain algorithm to prove copyright ownership and content integrity authentication of digital products. In recent years, image tampering watermark technology for information recovery has become a hot topic, such watermarking algorithm considers information that represents an image characteristic as watermark, then embed it into the image, The extracted watermark information is first used to localize the tampered region of the image, and then restore the tampered region.In this paper, we study on the fragile watermarking technology for image tamper detection, location and recovery. The details are as follows:1. An adaptive digital watermark embedding technology based on HVS (Human Visual System) is presented. Invisibility is an important characteristic of the watermarking algorithm, when a digital watermarking is embedded in image, it will not have a significant effect on the image quality, and the embedded watermarking information is not easily noticed.By utilizing the characteristics of the human visual system, we select the variance of the image region to determine the position of the adaptive watermark embedding, watermark will be embedded in the image areas of complex texture. By adaptive embedding, the proposed algorithm can reduce the impact of embedding digital watermark on the image as much as possible, and improve watermark invisibility.2. In order to authenticate the integrity of digital image content, a watermarking algorithm for positioning tampering area and restoring original image is presented. Embedded watermark is divided into random sequence for tamper detection and content watermark or image restoration. The content watermark consists of two parts, one part is obtained from the quantized average gray value of each2*2image block, which results in high accuracy of recovery while reducing the amount of watermark; another part is the average gray value of each4*4image block, it is then embedded in the original image via an adaptive way. Through this way, the decadency of the original image quality is small, and the accuracy of the restored image is also improved.The proposed algorithm is implemented via Matlab software, and the performance of the algorithm is analyzed through a large number of simulation experiments. Experimental results show that the algorithm has a good invisibility, the tampered regions can be accurately localized, and the quality of restored image can reach a high level. Compared with other algorithms, the proposed algorithm achieves better watermark invisibility, more accurate localization and better restored image quality.