Existing zero-watermarking algorithms for remote sensing images heavily rely on traditional feature extraction techniques, which are vulnerable to targeted attacks and lack discriminability for images captured by different sensors or at different time periods in the same geographical area. To address these limitations, this paper proposes a novel watermarking algorithm based on blockchain and Stacked Denoising Autoencoder (SDAE) to achieve lossless copyright protection for high-resolution remote sensing images (HRRS). The algorithm utilizes SDAE to extract deep and robust features from local square feature regions (LSFR) for watermark construction. Moreover, the algorithm incorporates a watermark registration scheme designed with Hyperledger Fabric and IPFS to ensure secure and trustworthy registration of watermarks and associated parameter information, enhancing the algorithm's uniqueness. Experimental results demonstrate the effectiveness of the proposed algorithm against various watermark attacks and its high discriminability for similar images. This algorithm holds significant potential for wide-ranging applications in the field of lossless copyright protection for HRRS, effectively safeguarding the commercial interests of data providers.
Open access
Advanced Steganography and Watermarking Techniques
In order to solve the problem that the existing zero-watermark technology relies on the third-party IPR management organization, this paper proposes a new zero- watermark algorithm based on blockchain. The algorithm considers the multi-band characteristics of remote sensing image. It uses K-L transform, NSCT transform, SVD and other technical means to obtain the robust zero-watermark image of remote sensing image. The zero-watermark is stored on the blockchain by a zero-watermark registration system based on IPFS and Hyperledger Fabric. The experimental results show that the algorithm based on blockchain proposed in this paper realizes the lossless protection of data copyright. Moreover, this paper proves that blockchain can replace third-party IPR management organization.
Advanced Image Fusion Techniques
Advanced Steganography and Watermarking Techniques
Tao Xiang, Honghong Zeng, Biwen Chen, Shangwei Guo
Medical image fusion generates a fused image containing multiple features extracted from different source images, and it is of great help in clinical analysis and diagnosis. However, training a deep learning model for image fusion usually requires enormous computing power, especially for large volumes of medical data. Meanwhile, the privacy of images is also a critical issue. In this article, we propose a privacy-preserving blockchain-based medical image fusion (BMIF) framework. First, to ensure fusion performance, we design a new medical image fusion model based on convolutional neural network and Inception network and integrate the proposed model into the consensus process of blockchain. Next, to save computing power of blockchain, we design a consensus mechanism by requesting consensus nodes to train the fusion model instead of calculating useless hash values in traditional blockchain. Then, to protect data privacy, we further present an efficient homomorphic encryption to realize the training of fusion model on encrypted medical data. Finally, we conduct theoretical analysis and extensive experiments on public datasets to evaluate the feasibility and the performance of our proposed BMIF. The results exhibit that BMIF is efficient and secure, and our medical image fusion network performs better than state-of-the-art approaches.
Computed tomography imaging spectrometry (CTIS) is a snapshot hyperspectral imaging technique that can obtain a three-dimensional (${2D +}\lambda$) data cube of the target scene within a single exposure. Previous studies of CTIS suggest that reconstructions usually suffer from severe artifacts due to the limited number of projections available. To overcome this limitation, an iterative algorithm combining superiorization and guided image filtering is proposed to explore the intrinsic properties of the hyperspectral data cube as well as the characteristics of zero-order diffraction for the first time, to the best of our knowledge. Results from both simulative studies and proof-of-concept experiments demonstrate its superiority in suppressing artifacts and improving precision over the frequently used expectation maximization algorithm.
This work aims to solve the problem of “big” images manipulation up to the order of gigapixels ones. The sequential elaboration of these images is too complex and many times, impossible. This is a problem present in many study areas like forensics investigation or medical analysis and there are few solutions in literature. In this work, we will develop a distributed ledger and Smart Contracts solution in order to manipulate super resolution images in a decentralized and competitive way. The proposed system is highly dependable as the Blockchain infrastructure guarantee fault tolerance, high security and reliability. In the second part of the work, the solution has been tested in the context of demographic analysis. Thanks to this work, we can develop a peer to peer computational power exchange where scientists are able to offer their own digital token in exchange of a Gigapixel image elaboration, classification and recognition.