This paper deals with the study and analysis of several rational approximations to approach the behavior of arbitrary-order differentiators and integrators in the frequency domain. From the Riemann–Liouville, Grünwald–Letnikov and Caputo basic definitions of arbitrary-order calculus until the reviewed approximation methods, each of them is coded in a Maple 18 environment and their behaviors are compared. For each approximation method, an application example is explained in detail. The advantages and disadvantages of each approximation method are discussed. Afterwards, two model order reduction methods are applied to each rational approximation and assist a posteriori during the synthesis process using analog electronic design or reconfigurable hardware. Examples for each reduction method are discussed, showing the drawbacks and benefits. To wrap up, this survey is very useful for beginners to get started quickly and learn arbitrary-order calculus and then to select and tune the best approximation method for a specific application in the frequency domain. Once the approximation method is selected and the rational transfer function is generated, the order can be reduced by applying a model order reduction method, with the target of facilitating the electronic synthesis.

Despite recent improvements in computer vision, artificial visual systems’ design is still daunting since an explanation of visual computing algorithms remains elusive. Salient object detection is one problem that is still open due to the difficulty of understanding the brain’s inner workings. Progress in this research area follows the traditional path of hand-made designs using neuroscience knowledge or, more recently, deep learning, a particular branch of machine learning. Recently, a different approach based on genetic programming appeared to enhance handcrafted techniques following two different strategies. The first method follows the idea of combining previous hand-made methods through genetic programming and fuzzy logic. The second approach improves the inner computational structures of basic hand-made models through artificial evolution. This research proposes expanding the artificial dorsal stream using a recent proposal based on symbolic learning to solve salient object detection problems following the second technique. This approach applies the fusion of visual saliency and image segmentation algorithms as a template. The proposed methodology discovers several critical structures in the template through artificial evolution. We present results on a benchmark designed by experts with outstanding results in an extensive comparison with the state of the art, including classical methods and deep learning approaches to highlight the importance of symbolic learning in visual saliency.

Advances in the development of collision-free path planning algorithms are the main need not only to solve mazes with robotic systems, but also for their use in modern product transportation or green logistics systems and planning merchandise deliveries inside or outside a factory. This challenge increases as the complexity of the task in its structure also increases. This paper deals with the development of a novel methodology for solving mazes with a mobile robot, using image processing techniques and graph theory. The novelty is that the mobile robot can find the shortest path from a start-point to the end-point into irregular mazes with abundant irregular obstacles, a situation that is not far from reality. Maze information is acquired from an image and depending on the size of the mobile robot, a grid of nodes with the same dimensions of the maze is built. Another contribution of this paper is that the size of the maze can be scaled from 1 m × 1 m to 66 m × 66 m, maintaining the essence of the proposed collision-free path planning methodology. Afterwards, graph theory is used to find the shortest path within the grid of reduced nodes due to the elimination of those nodes absorbed by the irregular obstacles. To avoid the mobile robot to travel through those nodes very close to obstacles and borders, resulting in a collision, each image of the obstacle and border is dilated taking into account the size of the mobile robot. The methodology was validated with two case studies with a mobile robot in different mazes. We emphasize that the maze solution is found in a single computational step, from the maze image as input until the generation of the Path vector. Experimental results show the usefulness of the proposed methodology, which can be used in applications such as intelligent traffic control, military, agriculture and so on.

This work shows the advantage of expert knowledge for leukemic cell recognition. In the medical area, visual analysis of microscopic images has regularly used biological samples to recognize hematological disorders. Nowadays, techniques of image recognition are needed to achieve an adequate identification of blood tissues. This paper presents a procedure to acquire expert knowledge from blood cell images. We apply Gaussian mixtures, evolutionary computing, and standard techniques of image processing to extract knowledge. This information feeds a support vector machine or multilayer perceptron to classify healthy or leukemic cells. Additionally, convolutional neural networks are used as a benchmark to compare our proposed method with the state of the art. We use a public database of 260 healthy and leukemic cell images. Results show that our traditional pattern recognition methodology matches deep learning accuracy since the recognition of blood cells achieves 99.63%, whereas the convolutional neural networks reach 97.74% on average. Moreover, the computational effort of our approach is minimal, while meeting the requirement of being explainable.