In this paper, we present deep learning-based blind image deblurring methods for estimating and removing a non-uniform motion blur from a single blurry image. We propose two fully convolutional neural networks (CNN) for solving the problem. The networks are trained end-to-end to reconstruct the latent sharp image directly from the given single blurry image without estimating and making any assumptions on the blur kernel, its uniformity, and noise. We demonstrate the performance of the proposed models and show that our approaches can effectively estimate and remove complex non-uniform motion blur from a single blurry image.
Information exchanged between two parties is often targeted by a third party. Steganography in images is often used for handling threats, such as, in this case, an attacker suspecting the existence of confidential information. But even in this case, if a pattern is used, e.g., hiding the information in successive or every third pixel, the attacker might discover the pattern, and the following revelation of the secret information will cause no difficulties. To avoid such vulnerabilities, the information is hidden in the frames of graphical animation. In the frames, the positions of the pixels selected to store confidential information in the given work are determined by the values of a mathematical function. These actions will ensure the high secrecy of information.
The paper presents the results of calculations and tests of the developed dataset expanding algorithm for training a generative-adversarial network. The research was conducted on two types of malicious software: mimikatz and cring. The boosting method was chosen as a method for expanding the database of datasets.
The process of expanding the database of datasets was carried out in a granular manner, using timestamps. Simulation of the algorithm operation at different iterations and visualization of the results have been carried out.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.