Humberto Sossa

National Polytechnic Institute | IPN · Robótica y Mecatrónica

In computer vision, human action recognition is a hot topic, popularized by the development of deep learning. Deep learning models typically accept video input without prior processing and train them to achieve recognition. However, conducting preliminary motion analysis can be beneficial in directing the model training to prioritize the motion of...

In computer vision, human action recognition is a hot topic, popularized by the development of deep learning. Current models have achieved high accuracy results on public datasets. Despite this success, they require significant computational resources for training. Given that transfer learning based techniques allow reusing what other models have a...

This work addresses the challenge of classifying multiclass visual EEG signals into 40 classes for brain–computer interface applications using deep learning architectures. The visual multiclass classification approach offers BCI applications a significant advantage since it allows the supervision of more than one BCI interaction, considering that e...

The Associative Pattern Classifier (APC) was designed as an associative memory, focusing particularly on pattern classification. This implies that the training memory is constructed in a single operation and pattern classification also occurs in a single process. It is important to note that the APC translates the input patterns through a translati...

The importance of sustainable cities and communities is increasingly critical in the quest to reverse environmental damage. During the COVID-19 pandemic and even today, disinfection methods were used to keep enclosed and crowded spaces constantly disinfected. In favor of sustainable development and public health destabilized by the COVID-19 pandemi...

Electricity is an essential energy resource in the industrial, commercial and housing sector, having a very important role in the development of societies. Urbanization and industrialization implies a great demand of energy for developing economies. In the search to be able to know how much electrical energy is consumed, a modeling of the electrica...

In this work, an evolutionary vision approach is used for the automatic recognition of AML leukemia images. Unlike common approaches using convolutional neural networks, in the presented model the feature extraction process is transparent. Moreover, the structure of the obtained solutions is amenable to interpretation by a human user, which is a si...

Specific anatomical structures from the female body, such as the axillary slope, armpit, pectoral muscle, or abdominal tissue, can be present in mammograms and might affect the proper mammogram analysis, especially in female populations with overweight issues, as is the case in Mexico. This work aims to determine if better results can be obtained i...

A Kalman filter can be used to fill space–state reconstruction dynamics based on knowledge of a system and partial measurements. However, its performance relies on accurate modeling of the system dynamics and a proper characterization of the uncertainties, which can be hard to obtain in real-life scenarios. In this work, we explore how the values o...

The quality of a city's infrastructure drives socioeconomic development. Specifically, it is important to streamline pavement quality monitoring to improve transportation. However, crack segmentation is a computational challenging problem that requires a fast response. In this paper, we propose a Fully Convolutional Network (FCN) for pavement crack...

In recent years, the recreational and commercial use of flight and driving simulators has become more popular. All these applications require the calculation of orientation in either two or three dimensions. Besides the Euler angles notation, other alternatives to represent rigid body rotations include axis-angle notation, homogeneous transformatio...

The computer vision community has proposed numerous formulations for object description based on human perceptivity and vast knowledge of the problem domain. In order to reduce human intervention, deep learning techniques are widely used to learn features automatically. However, they lack the property of explainability; that is, a human being under...

Neural cell counting is one of the ways in which damage caused by neurodegenerative diseases can be assessed, but it is not an easy task when it comes to neuronal counting in the most densely populated areas of the hippocampus. In this regard, this work presents a leveraged deep learning (DL) model, an innovative way to treat histological images an...

Noise management associated with input signals in sensor devices arises as one of the main problems limiting robot control performance. This article introduces a novel neuromorphic filter model based on a leaky integrate and fire (LIF) neural model cell, which encodes the primary information from a noisy input signal and delivers an output signal w...

Leukemia is a health problem that affects to world population causing thousands of kills yearly, thus accurate and human-readable diagnostic methods are required. Symbolic learning uses methods based on high-level representations of problems, which is useful to design interpretable models to understand the solutions found to solve a problem. In thi...

Neuromorphic computing is a recent class of brain-inspired high-performance computer platforms and algorithms involving biologically-inspired models adopting hardware implementation in integrated circuits. The neuromorphic computing applications have provoked the rise of highly connected neurons and synapses in analog circuit systems that can be us...

The Image Analysis community has widely used so-called bit-quads to propose formulations for computing the Euler characteristic of a 2-D binary image. Reported works have manually proposed different combinations of bit-quads to provide one or more formulations to calculate this important topological feature. This paper empirically shows how an Arti...

This paper describes the development of a hybrid system under a client-server architecture of multi-Tiers and logical multilayers, development with Nodejs-Express-Angular with Object Relational Mapping and Data Transfer Object with MariaDB.Generally, tracking systems with Global Navigation Satellite System technology are slow without a base archite...

In this chapter, we will provide the general and fundamental background related to two types of artificial neural networks techniques: spiking neural networks (SNN) and dendrite morphological neural networks (DMNN). The third generation of artificial neural networks, also known as SNN, has shown to be a very promising tool for the recognition of pa...

Brain-Computer Interfaces are new technologies with a fast development due to their possible usages, which still require overcoming some challenges to be readily usable. The paradigm of motor imagery is among the ones in these types of systems where the pipeline is tuned to work with only one person as it fails to classify the signals of a differen...

Dendrite morphological neurons (DMNs) are neural models for pattern classification, where dendrites are represented by a geometric shape enclosing patterns of the same class. This study evaluates the impact of three dendrite geometries--namely, box, ellipse, and sphere--on pattern classification. In addition, we propose using smooth maximum and min...

A large number of car accidents are caused by failures in the pavement. Their automatic detection is important for pavement maintenance, however, the current public datasets of images to train and test these systems contain a few hundred samples. In this paper, we introduce a new large dataset of images with more than 2000 samples that contains the...

Brain-computer interfaces are a promising technology for applications ranging from rehabilitation to video-games. A common problem for these systems is the ability to classify correctly signals corresponding to different subjects, as a consequence these systems are trained individually for each person. In this paper several classification methods,...

This paper presents an automatic method for obtaining formulas to calculate the Euler number in 2D binary images. This problem is addressed as a combinatorial optimization problem, where specific bit-quad patterns are optimally combined. An algorithm based on simulated annealing is devised to find optimal expressions to compute the Euler number, co...

This paper presents a method for pixel-wise classification applied for the first time on hippocampus histological images. The goal is achieved by representing pixels in a 14-D vector, composed of grey-level information and moment invariants. Then, several popular machine learning models are used to categorize them, and multiple metrics are computed...

An essential characteristic that an exploration robot must possess is to be autonomous. This is necessary because it will usually do its task in remote or hard-to-reach places. One of the primary elements of a navigation system is the information that can be acquired by the sensors of the environment in which it will operate. For this reason, an al...

Mobile robots integrate a combination of physical robotic elements for locomotion and artificial intelligence algorithms to move and explore the environment. They have the ability to react and make decisions based on the perception they receive from the environment to fulfill the assigned navigation tasks. A crucial issue in mobile robots is to add...

Traffic accidents represent one of the most serious problems around the world. Many efforts have been concentrated on implementing Advanced Driver Assistance Systems (ADAS) to increase safety by reducing critical tasks faced by the driver. In this paper, a Blind Spot Warning (BSW) system capable of virtualizing cars around the driver’s vehicle is p...

A method for performing 3D motion tracking of the shoulder joint with respect to the thorax, using MARG sensors and a data fusion algorithm, is proposed. Two tests were done: (1) qualitative and quantitative analysis of the response of the sensors, static position and during motion, with and without the proposed data fusion algorithm; (2) motion tr...

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 pro...

Breast cancer is a disease that has emerged as the second leading cause of cancer deaths in women worldwide. The annual mortality rate is estimated to continue growing. Cancer detection at an early stage could significantly reduce breast cancer death rates long-term. Many investigators have studied different breast diagnostic approaches, such as ma...

We study and propose, for the first time, an autonomous classification of the estrous cycle (the reproductive cycle in rats). This cycle consists of 4 stages: Proestrus, Estrus, Metestrus, and Diestrus. The short duration of the cycle in rats makes them an ideal model for research about changes that occur during the reproductive cycle. Classificati...

Building inspection is a vital task because infrastructure damage puts people at risk or causes economic losses. Thanks to the technological breakthroughs in regard to Unmanned Aerial Vehicles (UAVs) and intelligent systems, there is a real possibility to implement an inspection by means of these technologies. UAVs allow reaching difficult places a...

A three dimensional (3-D) digital image emerges as a straightforward extension of a two dimensional (2-D) digital image. A 3-D digital image can be obtained by digitizing the 3-D space in which one or more objects of interest can be contained. From each object in the digital image, several features describing their geometry and topology can be comp...

This data contains the Supplementary material for the paper "On the Accuracy and Computational Cost of Spiking Neuron Implementation." This is divided into five folders: 1) Source code, 2) Raw data, 3) Calculations, 4) Tables, and 5) Figures. The Source code folder has the script files written in Python 3.7.3. There are two sub-folders inside this...

Since more than a decade ago, three statements about spiking neuron (SN) implementations have been widely accepted: 1) Hodgkin and Huxley (HH) model is computationally prohibitive, 2) Izhikevich (IZH) artificial neuron is as efficient as Leaky Integrate-and-Fire (LIF) model, and 3) IZH model is more efficient than HH model (Izhikevich, 2004). As su...

Motivated by the recent progress of Spiking Neural Network (SNN) models in pattern recognition, we report on the development and evaluation of brain signal classifiers based on SNNs. The work shows the capabilities of this type of Spiking Neurons in the recognition of motor imagery tasks from EEG signals and compares their performance with other tr...

Two new hybrid neural architectures combining morphological neurons and perceptrons are introduced in this paper. The first architecture, called Morphological - Linear Neural Network (MLNN) consists of a hidden layer of morphological neurons and an output layer of classical perceptrons has the capability of extracting features. The second architect...

Dendrite morphological neurons are a type of artificial neural network that can be used to solve classification problems. The major difference with respect to classical perceptrons is that morphological neurons create hyperboxes to separate patterns from different classes, while perceptrons use hyperplanes. In this paper, we introduce an improved v...

A common approach to model kinematics for robot manipulators uses functions such as sin(), cos() and atan(),however, the rational trigonometry allows to exclude the use of these transcendental functions. Most of the computing processing to control a manipulator is dedicated to solving its kinematics, therefore, reduce the complexity of kinematic eq...

Pavement cracks are an increasing threat to public safety. Automatic pavement crack segmentation remains a very challenging problem due to crack texture inhomogeneity, high outlier potential, large variability of topologies, and so on. Due to this, automatic pavement crack detection has captured the attention of the computer vision community, and a...

In this paper we study the problem of rock detection in a Mars-like environment. We propose a convolutional neural network (CNN) to obtain a segmented image. The CNN is a modified version of the U-net architecture with a smaller number of parameters to improve the inference time. The performance of the methodology is proved in a dataset that contai...

Image enhancement techniques are needed to decrease the negative effects of blur or unwanted noise in image processing. In biomedical images, the quality of images is very important to achieve an adequate identification to detection or diagnosis purposes. This paper addresses the use of contrast enhancement to facilitate the identification of leuke...

We report the development and evaluation of brain signal classifiers, specifically Spiking Neuron based classifiers. The proposal consists of two main stages: feature extraction and pattern classification. The EEG signals used represent four motor imagery tasks: Left Hand, Right Hand, Foot and Tongue movements. In addition, one more class was added...

There are many applications in different fields, as diverse as computer graphics, medical imaging or pattern recognition for industries, where the use of three dimensional objects is needed. By the nature of these objects, it is very important to develop thrifty methods to represent, study and store them. In this paper, a new method to encode surfa...

Households are responsible for more than 40% of the global electricity consumption [7]. The analysis of this consumption to find unexpected behaviours could have a great impact on saving electricity. This research presents an experimental study of supervised and unsupervised neural networks for anomaly detection in electrical consumption. Multilaye...

A brain-computer interface provides individuals with a way to control a computer. However, most of these interfaces remain mostly utilized in research laboratories due to the absence of certainty and accuracy in the proposed systems. In this work, we acquired our own dataset from seven able-bodied subjects and used Deep Multi-Layer Perceptrons to c...

Dendrite morphological neural networks are emerging as an attractive alternative for pattern classification, providing competitive results with other classification methods. A key problem in the design of these neural networks is the election of the number of their dendrites. Most training methods are heuristics that do not optimize the learning pa...

Background The performance of Computer Aided Diagnosis Systems for early melanoma detection relies mainly on quantitative evaluation of the geometric features corresponding to skin lesions. In these systems, diagnosis is carried out by analyzing four geometric characteristics: asymmetry (A), border (B), color (C) and dimension (D). The main objecti...

Building inspection searching for superficial defects, such as cracks, is a vital task because such damages cause economic losses or put at risk the integrity of people. For this reason, different ways to reduce the costs and risks through the use of robotic systems that allow make inspections have been studied. Among these robotic systems, we have...

Brain–computer interfaces (BCI) rely on classification algorithms to detect the patterns of the brain signals that encode the mental task performed by the user. Therefore, robust and reliable classification techniques should be developed and evaluated to recognize the user's mental task with high accuracy. This paper proposes the use of the novel d...

Morphological neural networks, in particular, those with dendritic processing (MNNDPs), have shown to be a very promising tool for pattern classification. In this chapter, we present a survey of the most recent advances concerning MNNDPs. We provide the basics of each model and training algorithm; in some cases, we present simple examples to facili...

It is known that, depending on the numerical method, the simulation accuracy of a spiking neuron increases monotonically and that the computational cost increases in a power-law complexity as the time step reduces. Moreover, the accuracy and computational cost also are substantially affected by the mechanism responsible for generating the action po...

Precision Agriculture aims to apply selective treatments and tasks at localized areas concerning crop fields. Robotized and autonomous tractors, equipped with perception, decision-making and actuation systems, can apply specific treatments as may be required. Correct plant identification through the perception system, including crops and weeds, is...

Lane detection is a problem that has attracted in the last years the attention of the computer vision community. Most of approaches used until now to face this problem combine conventional image processing, image analysis and pattern classification techniques. In this paper, we propose a methodology based on so-called Ellipsoidal Neural Networks wi...

An algorithm that generates an exponential number of stable states for the very well-known Hopfield Neural Network (HNN) is introduced in this paper. We show that the quantity of stable states depends on the dimension and number of components of the input pattern supporting noise. Extensive tests verify that the states generated by our algorithm ar...

A new method to compute the Euler number of a 2-D binary image is described in this paper. The method employs three comparisons unlike other proposals that utilize more comparisons. We present two variations, one useful for the case of images containing only 4-connected objects and one useful in the case of 8-connected objects. To numerically valid...

Spiking neurons are mathematical models that simulate the generation of the electrical pulse at the neuron membrane. Most spiking neurons are expressed as a non-linear system of ordinary differential equations. Because these systems are hard to solve analytically, they must be solved using a numerical method through a discrete sequence of time step...

In previous works, a successful scheme using a single Spiking Neuron (SN) to solve complex problems in pattern recognition has been proposed. This consists in using the firing frequency response to classify a given input pattern, which is multiplied by a weight vector to produce a constant stimulation current. The weight vector is adjusted by an ev...

In this paper, a comparative study between two different neural network models is performed for a very simple type of classificaction problem in 2D. The first model is a deep neural network and the second is a dendrite morphological neuron. The metrics to be compared are: training time, classification accuracies and number of learning parameters. W...

Dendrite morphological neurons are a type of artificial neural network that works with min and max operators instead of algebraic products. These morphological operators build hyperboxes in N-dimensional space. These hyperboxes allow the proposal of training methods based on heuristics without using an optimisation method. In literature, it has bee...

We present two formulations and two procedures that can be used for computing the number of bubbles and tunnels of a 3-D binary object. The first formulation is useful to determine the number of bubbles of an object, while the second one can be used to calculate the number of tunnels of an object. Both formulations are formally demonstrated. Exampl...

The mean shift iterative algorithm (MSHi) was proposed in 2006 [1], where the entropy was used as a stopping criterion. This algorithm was employed to carry out image segmentation. From then on, a theoretical base has been developed and a group of applications has been carried out using this algorithm. This paper proposes a new stopping criterion f...