Search

In this article, an optical image encryption algorithm based on chaotic baker’s map is presented. The stego-image is encrypted with the help of double random phase encoding algorithm and then produced disorder with the help of chaotic transformation. Security test shows that the reading of proposed algorithm is very close to the optimal values.

Mathematical models can produce desired dynamics and statistical properties with the insertion of suitable nonlinear terms, while energy characteristics are crucial for practical application because any hardware realizations of nonlinear systems are relative to energy flow. The involvement of memristive terms relative to memristors enables multistability and initial-dependent property in memristive systems. In this study, two kinds of memristors are used to couple a capacitor or an inductor, along with a nonlinear resistor, to build different neural circuits. The corresponding circuit equations are derived to develop two different types of memristive oscillators, which are further converted into two kinds of memristive maps after linear transformation. The Hamilton energy function for memristive oscillators is obtained by applying the Helmholz theorem or by mapping from the field energy of the memristive circuits. The Hamilton energy functions for both memristive maps are obtained by replacing the gains and discrete variables for the memristive oscillator with the corresponding parameters and variables. The two memristive maps have rich dynamic behaviors including coherence resonance under noisy excitation, and an adaptive growth law for parameters is presented to express the self-adaptive property of the memristive maps. A digital signal process (DSP) platform is used to verify these results. Our scheme will provide a theoretical basis and experimental guidance for oscillator-to-map transformation and discrete map-energy calculation.

The rising number and size of remote sensing (RS) image archives makes content-based RS image retrieval (CBRSIR) more important. Convolutional neural networks (CNNs) offer good CBRSIR performance, but the features they extract are not rotation-invariant. This is problematic as objects in RS images appear in arbitrary rotation angles. We develop and investigate two new rotation-aware CNN-based CBRSIR methods: 1) In the Feature Map Transformation Based Rotation-Aware Network (FMT-RAN), the last pooling layer is rotated in four different angles during training. Its outputs are passed through the same fully connected-, coding-, and classification layer, and the resulting losses are added. 2) The Spatial Transformer-based Rotation-Aware Network (ST-RAN) contains a spatial transformer network (STN) and a rotation aware network (RAN). For training, the original and a randomly rotated version of an image are fed into the ST-RAN. The STN generates a transformed version of the original to match the rotated image. The RAN extracts the features of all three images. We apply two-stage training, which first optimizes the STN and then the RAN. Both of our methods are efficient in terms of retrieval accuracy and time, but ST-RAN has the overall best performance. It outperforms the state-of-the-art CBRSIR methods. [ABSTRACT FROM AUTHOR]

A map transformation is a willful geometrical distortion of an image. Examples are rotation, shearing, and local stretching of the coordinate system. We describe an optical setup that performs a map transformation by spatial filtering with two phase-only filters. The system is able to perform in x- and y-linear distortions (e.g., shears) as well as certain x- and y-nonlinear distortions. The distorting filters introduce no aberrations. The object may radiate coherently, incoherently, or partially coherently. Some experimental results are presented.

The human visual system can recognize object categories accurately and efficiently and is robust to complex textures and noises. To mimic the analogy-detail dual-pathway human visual cognitive mechanism revealed in recent cognitive science studies, in this article, we propose a novel convolutional neural network (CNN) architecture named analogy-detail networks (ADNets) for accurate object recognition. ADNets disentangle the visual information and process them separately using two pathways: the analogy pathway extracts coarse and global features representing the gist (i.e., shape and topology) of the object, while the detail pathway extracts fine and local features representing the details (i.e., texture and edges) for determining object categories. We modularize the architecture and encapsulate the two pathways into the analogy-detail block as the CNN building block to construct ADNets. For implementation, we propose a general principle that transmutes typical CNN structures into the ADNet architecture and applies the transmutation on representative baseline CNNs. Extensive experiments on CIFAR10, CIFAR100, street view house numbers, and ImageNet data sets demonstrate that ADNets significantly reduce the test error rates of the baseline CNNs by up to 5.76% and outperform other state-of-the-art architectures. Comprehensive analysis and visualizations further demonstrate that ADNets are interpretable and have a better shape-texture tradeoff for recognizing the objects with complex textures. [ABSTRACT FROM AUTHOR]

The article informs about a cartogram depicting tracks of tropical storms from 1945 to 2008 and thus helps in visualization of global tropical storm tracks. It reports on the increase in number of tropical cyclones and tropical cyclone intensity and their impact of populations live in coastal regions due to landfall caused by them. It informs about the regions vulnerable to tropical storms such as Caribbean Islands and Bay of Bengal.