We analyzed the effects of global pooling measurements on extracting relevant texture information from a given set of activation maps. Initially, using a layer-by-layer approach (GP-CNN), we experimentally demonstrated that layers at various depth levels could provide high-quality texture information. Based on this finding, we developed RankGP-CNN, a method that performs multi-layer feature extraction. More specifically, RankGP-CNN treats every CNN model as a vast collection of deep composite functions, where each function computes a 2D activation map for every input image. A feature ranking approach then assigns a score to each deep composite function by processing the activation maps generated for a particular dataset bank. Eventually, RankGP-CNN uses the top-ranked deep composite functions to compute feature vectors for different texture datasets. Experiments on a dedicated classifier showed that RankGP-CNN achieves good results and can adapt to different texture problems. Finally, we present RankGP-3M-CNN as the version of RankGP-CNN that considers multiple CNN models. Overall, RankGP-3M-CNN achieves promising results with the advantage of only using the default scale of the input images.

The development of new pseudo-random number generators (PRNGs) has steadily increased over the years. Commonly, PRNGs' randomness is "measured" by using statistical pass/fail suite tests, but the question remains, which PRNG is the best when compared to others. Existing randomness tests lack means for comparisons between PRNGs, since they are not quantitatively analysing. It is, therefore, an important task to analyze the quality of randomness for each PRNG, or, in general, comparing the randomness property among PRNGs. In this paper, we propose a novel visual approach to analyze PRNGs randomness allowing for a ranking comparison concerning the PRNGs' quality. Our analysis approach is applied to ensembles of time series which are outcomes of different PRNG runs. The ensembles are generated by using a single PRNG method with different parameter settings or by using different PRNG methods. We propose a similarity metric for PRNG time series for randomness and apply it within an interactive visual approach for analyzing similarities of PRNG time series and relating them to an optimal result of perfect randomness. The interactive analysis leads to an unsupervised classification, from which respective conclusions about the impact of the PRNGs' parameters or rankings of PRNGs on randomness are derived. We report new findings using our approach in a study of randomness for state-of-the-art numerical PRNGs such as LCG, PCG, SplitMix, Mersenne Twister, and RANDU as well as chaos-based PRNG families such as K-Logistic map and K-Tent map with varying parameter K.

We propose a descriptor based on binary patterns extracted from network-automata time-evolution patterns (TEP) aiming to characterize networks. More, in particular, we explore TEPs descriptors from the Life-Like Network Automata (LLNA), a cellular automaton inspired by the rules of the "Life-Like" family that uses a network as tessellation, and based on its dynamics to extract features for network characterization. In recent work, the LLNA has been introduced as a pattern recognition tool that uses a descriptor based on the histograms of complexity measures such as the entropy, word length, and Lempel-Ziv complexity. However, these descriptors correspond to continuous values, and consequently, their histograms lack of an optimal number of bins, which therefore turns out to be a parametric issue. To overcome this disadvantage, we propose a new descriptor that computes feature vectors formed by discrete binary patterns histograms with different lengths D. Furthermore, we show a statistical improvement of the proposed method compared to earlier approaches such as the original LLNA and classical network structural measurements. Our experimental results show the performance improvement of the proposed method in six synthetic network databases and eight real network databases.

Texture is one of the most-studied visual attribute for image characterization since the 1960s. However, most hand-crafted descriptors are monochromatic, focusing on the gray scale images and discarding the color information. In this context, this work focus on a new method for color texture analysis considering all color channels in a more intrinsic approach. Our proposal consists of modeling color images as directed complex networks that we named Spatio-Spectral Network (SSN). Its topology includes within-channel edges that cover spatial patterns throughout individual image color channels, while between-channel edges tackle spectral properties of channel pairs in an opponent fashion. Image descriptors are obtained through a concise topological characterization of the modeled network in a multiscale approach with radially symmetric neighborhoods. Experiments with four datasets cover several aspects of color-texture analysis, and results demonstrate that SSN overcomes all the compared literature methods, including known deep convolutional networks, and also has the most stable performance between datasets, achieving $98.5(\pm1.1)$ of average accuracy against $97.1(\pm1.3)$ of MCND and $96.8(\pm3.2)$ of AlexNet. Additionally, an experiment verifies the performance of the methods under different color spaces, where results show that SSN also has higher performance and robustness.

This work proposes a method to extract features from texture images by applying a Gaussian pyramid multiscale approach to the Bouligand–Minkowski fractal descriptors. The proposal starts from the texture image and computes the stack of multi-resolution images that compose the pyramid, in both directions, of reduction and expansion. In the following, each image in the stack is mapped onto a surface, which is dilated by spheres with variable radii and the dilation volumes are used to compute the Bouligand–Minkowski fractal descriptors for each level. Both the descriptors of each level and combinations with descriptors from the original image are verified in the classification of well-known databases of textural images. The proposed method outperformed other classical and state-of-the-art descriptors with a significant advantage in most cases, including situations where random noise is added to the images.

This work presents a new procedure to extract features of grey-level texture images based on the discrete Schroedinger transform. This is a non-linear transform where the image is mapped as the initial probability distribution of a wave function and such distribution evolves in time following the Schroedinger equation from Quantum Mechanics. The features are provided by standard deviation of the distribution measured at different times. The proposed method is applied to the classification of three databases of textures used for benchmark and compared to other well-known texture descriptors in the literature, such as textons, local binary patterns, multifractals, among others. All of them are outperformed by the proposed method in terms of percentage of images correctly classified. The proposal is also applied to the identification of plant species using scanned images of leaves and again it outperforms other texture methods. A test with images affected by Gaussian and "salt & pepper" noise is also carried out, also with the best performance achieved by the Schroedinger descriptors.

This work presents a method for color texture analysis based on fractal geometry. The method is based on its predecessor [4] and consists of mapping each color channel onto a surface and dilating such surface by spheres with a variable radius. The descriptors are obtained from the relation between the volumes of the dilated surfaces and the dilation radii. The dilation process creates a mutual interference among the color channels. The proposed descriptors measure the degree of such interference as well as the complexity of pixel intensity arrangements. This combination provides a robust and precise texture description. The efficiency of the method is assessed in a classification task of well-known texture data sets and the results demonstrate that it outperforms the best approaches described in the literature.

Much work has been done in the field of texture analysis and classification. While promising classification methods have been proposed, most of them rely on classical image analysis approaches. This paper presents a texture classification method based on diffusion in directed networks. First, an image is modeled as a directed network by mapping each pixel as a node and connecting two nodes up to a maximum distance r. To reveal texture properties, links between two nodes are removed based on the pixel intensity difference. Once such a network is obtained, the activity of each node is estimated by random walks and combined into a histogram to describe the image. The main contribution of this paper is the use of directed networks, which tends to provide better performance than in undirected cases. Also, we have shown that the activity induced on these networks can be effectively used as texture descriptor. Experimental results show that the proposed method is favorably compared to traditional texture methods on widely used texture datasets. The proposed method is also found to be promising for plant species classification using samples of leaf texture.

The spatio-temporal dynamics of cellular automata (CAs) has attracted the attention of researchers from different fields, mainly mathematics, computer science and engineering, as a consequence of both the intriguing spatio-temporal patterns that these dynamical systems evolve and the fact that they enable the modelling of complex natural phenomena. Yet, to this day, there are only a few studies that focus on the automated classification of cellular automata on the basis of the space-time diagrams they evolve. Here, we present an innovative approach to classify CAs according to Wolfram's classification scheme in an automated way by relying on texture descriptors that capture the nature of the evolved space-time diagrams. More specifically, we propose the use of one of two well-known texture descriptors, namely Local Binary Pattern Variance and Fourier descriptors, to generate features grasping the diagrams' nature, followed by nearest neighbor classification. The performance of this approach is assessed through a cross-validation and by analysing the percentage of pre-classified rules that is required to arrive at an acceptable success rate. The experiments involve the family of elementary CAs and four families of totalistic CAs with neighborhood radii ranging from one to three, and a state space consisting of up to three states. The results show the potential of our proposal with success rates varying between 65% and 98% depending on the size of the training set, which ranges from 10% to 90% of the rules in the CA family at stake. For totalistic CAs, this training set should be classified manually to start the process of automated classification.