This paper presents a high discriminative texture analysis method based on the fusion of complex networks and randomized neural networks. In this approach, the input image is modeled as a complex networks and its topological properties as well as the image pixels are used to train randomized neural networks in order to create a signature that represents the deep characteristics of the texture. The results obtained surpassed the accuracies of many methods available in the literature. This performance demonstrates that our proposed approach opens a promising source of research, which consists of exploring the synergy of neural networks and complex networks in the texture analysis field.

Dynamic texture analysis has been the focus of intensive research in recent years. Thus, this paper presents an innovative and highly discriminative dynamic texture analysis method, whose signature is composed of the weights of the output layer of a randomized neural network after a training procedure. This training is performed by using the pixels of slices of each orthogonal plane of the video (XY, YT, and XT) as input feature vectors and corresponding output labels. The obtained video signature provided an accuracy of 97.05%, 98.54%, 97.74% and 96.51% on the UCLA-50 classes, UCLA-9 classes, UCLA-8 classes and Dyntex++, respectively. These results, when compared to other dynamic texture analysis methods, demonstrate that our descriptors are very effective and that our proposed approach can contribute significantly to the field of dynamic texture analysis.

Shape analysis is a very important field in computer vision. This work presents a novel and highly discriminative shape analysis method based on the weights of a Randomized Neural Network (RNN). Two approaches are proposed to extract the contour signature: Neighborhood approach uses the distance of each contour pixel and its immediate neighboring pixels and Contour portion approach, which uses metrics computed from contour sections to model the shape as RNN. We also proposed a signature that combines the feature vectors resulting from both approaches, thus resulting in a set of features tolerant to affine transformations, such as rotation and scale. We compared our approach with other shape analysis methods in 6 different shapes datasets. We calculated the accuracy as measure performance and obtained 97.98%, 99.07%, 84.67%, 87.67%, 88.92% and 80.58% for Kimia, Fish, Leaf, Rotated Leaf, Scaled Leaf and Noised Leaf datasets, respectively. The achieved performance of our method surpassed the results of several compared methods in most of these datasets, thus proving that our proposed signature can be applied successfully in shape analysis problems.

Classification of pap-smear cell images is a relevant and challenging medical problem. In order to contribute to the analysis of these images, we propose to compute complexity based features from each color channel of them. Moreover, we propose to use the gravitational method, a novel and very discriminative approach which improves our ability to extract descriptors from the images. We compared our approach with results obtained from other color texture analysis methods by using both accuracy and AUC, which are measurements of performance. The obtained results (89.64% for accuracy and 0.9086 for AUC) demonstrate that color texture classification using the gravitational model is a feasible approach for the classification of pap-smear images.

Color textures are among the most important visual attributes in image analysis. This paper presents a novel method to analyze color textures by modeling a color image as a graph in two different and complementary manners (each color channel separately and the three color channels altogether) and by obtaining statistical moments from the shortest paths between specific vertices of this graph. Such an approach allows to create a set of feature vectors, which were extracted from VisTex, USPTex, and TC00013 color texture databases. The best classification results were 99.07%, 96.85%, and 91.54% (LDA with leave-one-out), 87.62%, 66.71%, and 88.06% (1NN with holdout), and 98.62%, 96.16%, and 91.34% (LDA with holdout) of success rate (percentage of samples correctly classified) for these three databases, respectively. These results prove that the proposed approach is a powerful tool for color texture analysis to be explored.