Distributed redundant image storages and reconstruction algorithm to contents verification

In this paper, we propose a novel contents verification method using the distributed redundant image storages and user-based reconstruction algorithm. There are two processes composing in our proposal such as the image dispersion method and the image reconstruction method. In the image dispersion method, the images are dispersed into many information patches and are stored on the distributed storages, e.g Dropbox, Google Drive, ... Since the size of information patches is smaller than the full-size of image, the information patches can be distributed on cloud storages via network. Therefore, the resource of cloud storages is used effectively. On the other hand, in the image reconstruction method, the information patches of the image can be collected in order to reconstruct the image for user. The key point of idea in the image reconstruction method is that the reconstructed images should be distinguished each others and the producer can detect the user that it belongs to. The algorithm of reconstruction method employs the watermarking method to generate the watermark pattern based on the user's information such as license number. The demonstration results of our proposed method show that our system can be used as the contents verification in the digital contents distribution system.