Digital watermarking has been widely studied as a method of protecting the intellectual property rights of digital images, which are high value-added contents. Recently, studies implementing these techniques with neural networks have been conducted. This paper also proposes a neural network to perform a robust, invisible blind watermarking for digital images. It is a convolutional neural network (CNN)-based scheme that consists of pre-processing networks for both host image and watermark, a watermark embedding network, an attack simulation for training, and a watermark extraction network to extract watermark whenever necessary. It has three peculiarities for the application aspect: The first is the host image resolution’s adaptability. This is to apply the proposed method to any resolution of the host image and is performed by composing the network without using any resolution-dependent layer or component. The second peculiarity is the adaptability of the watermark information. This is to provide usability of any user-defined watermark data. It is conducted by using random binary data as the watermark and is changed each iteration during training. The last peculiarity is the controllability of the trade-off relationship between watermark invisibility and robustness against attacks, which provides applicability for different applications requiring different invisibility and robustness. For this, a strength scaling factor for watermark information is applied. Besides, it has the following structural or in-training peculiarities. First, the proposed network is as simple as the most profound path consists of only 13 CNN layers, which is through the pre-processing network, embedding network, and extraction network. The second is that it maintains the host’s resolution by increasing the resolution of a watermark in the watermark pre-processing network, which is to increases the invisibility of the watermark. Also, the average pooling is used in the watermark pre-processing network to properly combine the binary value of the watermark data with the host image, and it also increases the invisibility of the watermark. Finally, as the loss function, the extractor uses mean absolute error (MAE), while the embedding network uses mean square error (MSE). Because the extracted watermark information consists of binary values, the MAE between the extracted watermark and the original one is more suitable for balanced training between the embedder and the extractor. The proposed network’s performance is confirmed through training and evaluation that the proposed method has high invisibility for the watermark (WM) and high robustness against various pixel-value change attacks and geometric attacks. Each of the three peculiarities of this scheme is shown to work well with the experimental results. Besides, it is exhibited that the proposed scheme shows good performance compared to the previous methods.

We propose a new learning and inferring model that generates digital holograms using deep neural networks (DNNs). This DNN uses a generative adversarial network, trained to infer a complex two-dimensional fringe pattern from a single object point. The intensity and fringe patterns inferred for each object point were multiplied, and all the fringe patterns were accumulated to generate a perfect hologram. This method can achieve generality by recording holograms for two spaces (16 Space and 32 Space). The reconstruction results of both spaces proved to be almost the same as numerical computer-generated holograms by showing the performance at 44.56 and 35.11 dB, respectively. Through displaying the generated hologram in the optical equipment, we proved that the holograms generated by the proposed DNN can be optically reconstructed.

In this paper, we propose a new hardware architecture implemented as a very large scaled integrated circuit by using an application-specific integrated circuit technology, where block-based calculations are used to generate holograms. The proposed hardware is structured to produce a part of a hologram in the block units in parallel. A block of a hologram is calculated using an object point, and then the calculation is repeated for all object points to obtain intermediate results that are accumulated to produce the final block of a hologram. This structure can be used to produce holograms of various sizes in real time with optimized memory access. The proposed hardware was implemented using the Hynix 0.18 μm CMOS technology of Magna Chip, Inc., and it has about 448 K gate counts and a silicon size of $3.592\mathrm{mm}\times 3.592\mathrm{mm}$. It can generate complex holograms and operate in a stable manner at a clock frequency of 200 MHz.

This paper proposes a new adaptive watermarking scheme for digital images, which has the properties of blind extraction, invisibility, and robustness against attacks. The typical scheme for invisibility and robustness consisted of two main techniques: finding local positions to be watermarked and mixing or embedding the watermark into the pixels of the locations. In finding the location, however, our scheme uses a global space such that the multiple watermarking data is spread out over all four lowest-frequency subbands, resulting from n-level Mallat-tree 2D (dimensional) DWT, where n depends on the amount of watermarking data and the resolution of the host image, without any further process to find the watermarking locations. To embed the watermark data into the subband coefficients, weighting factors are used according to the type and energy of each subband to adjust the strength of the watermark, so we call this an adaptive scheme. To examine the ability of the proposed scheme, images with various resolutions are tested for various attacks, both pixel-value changing attacks and geometric attacks. With experimental results and comparison to the existing works we show that the proposed scheme has better performance than the previous works, except those which specialize in certain types of attacks.