Image steganalysis methods aim at detecting whether there exist hidden messages in images. Deep learning (DL) models have been proposed to enhance steganography detection. These models occupy a large amount of memory and, for this reason, should be optimized when the scenario involves resource-limited devices and systems. This work addresses different deep learning model optimization strategies, namely model pruning, quantization and weight clustering, applied to a deep learning model that presents competitive accuracy results in image steganalysis and belongs to the family of DL models with smaller memory requirements. The results show that the use of optimization schemes can lead to similar or even better accuracy compared to the original model (without the use of optimization schemes), while requiring less memory to store the model. Different scenarios are simulated for each optimization technique, and, finally, quantization is combined with pruning. For dynamic range quantization (DRQ), we achieve models that can save approximately 72% of storage. For FP16 quantization, we obtain better accuracy results and a model with approximately 50% less memory consumption. By applying weight clustering, we also achieve compressed models that can save more than 72% of storage space and lead to better accuracy for some scenarios. Using the combination of pruning and quantization, smaller models in terms of memory requirements are obtained.

This article aims to review encryption techniques for 3D or higher-dimensional images. Precisely, the following classes of images are considered: light field images, point clouds, and 3D medical images. The security and performance aspects of the encryption schemes for images are analyzed. It is observed that the number of papers with encryption in the transform domain exceeds the number of papers with encryption in the spatial domain. Additionally, it is observed that the application of encryption does not predominate in all the dimensions of the images, that is, the encryption does not occur completely in the image but rather through slices or sections. This review article summarizes several issues related to encryption techniques, technical perspectives for the future, and gaps in the literature.

Ocultação de informação é um tema que tem despertado o interesse de muitos pesquisadores, tendo sido abordado em dissertações de mestrado, teses de doutorado e em disciplinas de alguns cursos de gradução e de pós-graduação das áreas de Engenharia Elétrica e de Ciência da Computação. A ocultação de informação abrange áreas como esteganografia e marca d’água. Há um amplo espectro de aplicações associadas a essas técnicas, que vão desde comunicações secretas até direitos autorais (copyright) de objetos digitais. Este artigo apresenta fundamentos básicos de esteganografia e marca d’água em imagens digitais. São abordadas técnicas de esteganálise, bem como aplicações e aspectos de marcas d’água frágeis e robustas. É apresentado um breve levantamento de ferramentas que podem ser usadas como apoio ao ensino dessas técnicas. O artigo apresenta, particularmente, cenários em que ficam claros os benefícios das ferramentas para os estudantes.