Gonzalo Farias | Universidad de Chile (original) (raw)

Papers by Gonzalo Farias

Research paper thumbnail of Detection of Human Gait Phases Using Textile Pressure Sensors: A Low Cost and Pervasive Approach

Sensors

Human gait analysis is a standard method used for detecting and diagnosing diseases associated wi... more Human gait analysis is a standard method used for detecting and diagnosing diseases associated with gait disorders. Wearable technologies, due to their low costs and high portability, are increasingly being used in gait and other medical analyses. This paper evaluates the use of low-cost homemade textile pressure sensors to recognize gait phases. Ten sensors were integrated into stretch pants, achieving an inexpensive and pervasive solution. Nevertheless, such a simple fabrication process leads to significant sensitivity variability among sensors, hindering their adoption in precision-demanding medical applications. To tackle this issue, we evaluated the textile sensors for the classification of gait phases over three machine learning algorithms for time-series signals, namely, random forest (RF), time series forest (TSF), and multi-representation sequence learner (Mr-SEQL). Training and testing signals were generated from participants wearing the sensing pants in a test run under l...

Research paper thumbnail of Thermal Face Generation Using StyleGAN

IEEE Access, 2021

This article proposes the use of generative adversarial networks (GANs) via StyleGAN2 to create h... more This article proposes the use of generative adversarial networks (GANs) via StyleGAN2 to create high-quality synthetic thermal images and obtain training data to build thermal face recognition models using deep learning. We employed different variants of StyleGAN2, incorporating the new improved version of StyleGAN that uses adaptive discriminator augmentation (ADA). In addition, three different thermal databases from the literature were employed to train a thermal face detector based on YOLOv3 and to train StyleGAN2 and its variants, evaluating different metrics. The synthetic thermal database was built using GANSpace to manipulate the intermediate latent space w of StyleGAN2 and obtain images with different characteristics, such as eyeglasses, rotation, beards, etc. We carried out the training of 6 pretrained deep learning models for face recognition to validate the use of our synthetic thermal database, obtaining 99.98% accuracy for classifying synthetic thermal face images. INDEX TERMS Generative adversarial networks, StyleGAN2, thermal face recognition, deep learning.

Research paper thumbnail of Position control of a mobile robot using reinforcement learning

Research paper thumbnail of Online Virtual Control Laboratory of Mobile Robots

Research paper thumbnail of Face Recognition and Drunk Classification Using Infrared Face Images

Journal of Sensors, 2018

The aim of this study is to propose a system that is capable of recognising the identity of a per... more The aim of this study is to propose a system that is capable of recognising the identity of a person, indicating whether the person is drunk using only information extracted from thermal face images. The proposed system is divided into two stages, face recognition and classification. In the face recognition stage, test images are recognised using robust face recognition algorithms: Weber local descriptor (WLD) and local binary pattern (LBP). The classification stage uses Fisher linear discriminant to reduce the dimensionality of the features, and those features are classified using a classifier based on a Gaussian mixture model, creating a classification space for each person, extending the state-of-the-art concept of a “DrunkSpace Classifier.” The system was validated using a new drunk person database, which was specially designed for this work. The main results show that the performance of the face recognition stage was 100% with both algorithms, while the drunk identification saw...

Research paper thumbnail of A Neural Network Approach for Building An Obstacle Detection Model by Fusion of Proximity Sensors Data

Sensors (Basel, Switzerland), Jan 25, 2018

Proximity sensors are broadly used in mobile robots for obstacle detection. The traditional calib... more Proximity sensors are broadly used in mobile robots for obstacle detection. The traditional calibration process of this kind of sensor could be a time-consuming task because it is usually done by identification in a manual and repetitive way. The resulting obstacles detection models are usually nonlinear functions that can be different for each proximity sensor attached to the robot. In addition, the model is highly dependent on the type of sensor (e.g., ultrasonic or infrared), on changes in light intensity, and on the properties of the obstacle such as shape, colour, and surface texture, among others. That is why in some situations it could be useful to gather all the measurements provided by different kinds of sensor in order to build a unique model that estimates the distances to the obstacles around the robot. This paper presents a novel approach to get an obstacles detection model based on the fusion of sensors data and automatic calibration by using artificial neural networks.

Research paper thumbnail of Detection of Human Gait Phases Using Textile Pressure Sensors: A Low Cost and Pervasive Approach

Sensors

Human gait analysis is a standard method used for detecting and diagnosing diseases associated wi... more Human gait analysis is a standard method used for detecting and diagnosing diseases associated with gait disorders. Wearable technologies, due to their low costs and high portability, are increasingly being used in gait and other medical analyses. This paper evaluates the use of low-cost homemade textile pressure sensors to recognize gait phases. Ten sensors were integrated into stretch pants, achieving an inexpensive and pervasive solution. Nevertheless, such a simple fabrication process leads to significant sensitivity variability among sensors, hindering their adoption in precision-demanding medical applications. To tackle this issue, we evaluated the textile sensors for the classification of gait phases over three machine learning algorithms for time-series signals, namely, random forest (RF), time series forest (TSF), and multi-representation sequence learner (Mr-SEQL). Training and testing signals were generated from participants wearing the sensing pants in a test run under l...

Research paper thumbnail of Thermal Face Generation Using StyleGAN

IEEE Access, 2021

This article proposes the use of generative adversarial networks (GANs) via StyleGAN2 to create h... more This article proposes the use of generative adversarial networks (GANs) via StyleGAN2 to create high-quality synthetic thermal images and obtain training data to build thermal face recognition models using deep learning. We employed different variants of StyleGAN2, incorporating the new improved version of StyleGAN that uses adaptive discriminator augmentation (ADA). In addition, three different thermal databases from the literature were employed to train a thermal face detector based on YOLOv3 and to train StyleGAN2 and its variants, evaluating different metrics. The synthetic thermal database was built using GANSpace to manipulate the intermediate latent space w of StyleGAN2 and obtain images with different characteristics, such as eyeglasses, rotation, beards, etc. We carried out the training of 6 pretrained deep learning models for face recognition to validate the use of our synthetic thermal database, obtaining 99.98% accuracy for classifying synthetic thermal face images. INDEX TERMS Generative adversarial networks, StyleGAN2, thermal face recognition, deep learning.

Research paper thumbnail of Position control of a mobile robot using reinforcement learning

Research paper thumbnail of Online Virtual Control Laboratory of Mobile Robots

Research paper thumbnail of Face Recognition and Drunk Classification Using Infrared Face Images

Journal of Sensors, 2018

The aim of this study is to propose a system that is capable of recognising the identity of a per... more The aim of this study is to propose a system that is capable of recognising the identity of a person, indicating whether the person is drunk using only information extracted from thermal face images. The proposed system is divided into two stages, face recognition and classification. In the face recognition stage, test images are recognised using robust face recognition algorithms: Weber local descriptor (WLD) and local binary pattern (LBP). The classification stage uses Fisher linear discriminant to reduce the dimensionality of the features, and those features are classified using a classifier based on a Gaussian mixture model, creating a classification space for each person, extending the state-of-the-art concept of a “DrunkSpace Classifier.” The system was validated using a new drunk person database, which was specially designed for this work. The main results show that the performance of the face recognition stage was 100% with both algorithms, while the drunk identification saw...

Research paper thumbnail of Study of Visible Face Recognition Methods Applied to Infrared Spectrum

6th Chilean Conference on Pattern Recognition (CCPR), 2014

Research paper thumbnail of A Neural Network Approach for Building An Obstacle Detection Model by Fusion of Proximity Sensors Data

Sensors (Basel, Switzerland), Jan 25, 2018

Proximity sensors are broadly used in mobile robots for obstacle detection. The traditional calib... more Proximity sensors are broadly used in mobile robots for obstacle detection. The traditional calibration process of this kind of sensor could be a time-consuming task because it is usually done by identification in a manual and repetitive way. The resulting obstacles detection models are usually nonlinear functions that can be different for each proximity sensor attached to the robot. In addition, the model is highly dependent on the type of sensor (e.g., ultrasonic or infrared), on changes in light intensity, and on the properties of the obstacle such as shape, colour, and surface texture, among others. That is why in some situations it could be useful to gather all the measurements provided by different kinds of sensor in order to build a unique model that estimates the distances to the obstacles around the robot. This paper presents a novel approach to get an obstacles detection model based on the fusion of sensors data and automatic calibration by using artificial neural networks.

Research paper thumbnail of Study of Visible Face Recognition Methods Applied to Infrared Spectrum

6th Chilean Conference on Pattern Recognition (CCPR), 2014

Research paper thumbnail of Thermal face recognition over time using sparse representation approach

6th Chilean Conference on Pattern Recognition (CCPR), 2014

Research paper thumbnail of Thermal face recognition over time using sparse representation approach

6th Chilean Conference on Pattern Recognition (CCPR), 2014

Research paper thumbnail of Fusion of Visible and Thermal Descriptors Using Genetic Algorithms for Face Recognition Systems

Research paper thumbnail of Laboratorio de prácticas para la enseñanza de sistemas de control de tiempo real

Resumen: El presente trabajo describe el diseño y desarrollo de un laboratorio de prácticas espec... more Resumen: El presente trabajo describe el diseño y desarrollo de un laboratorio de prácticas especialmente concebido para apoyar el aprendizaje en sistemas de control de tiempo real. La herramienta desarrollada permite realizar experiencias de control de tiempo real sobre un motor de corriente continua tanto en modo simulación (basado en el modelo del proceso) o bien realizando pruebas prácticas usando el motor físico. Así, mediante la asignación de periodos de muestreo, tiempos de cómputo de tareas y prioridades de ejecución, el usuario final de la aplicación (profesores y estudiantes) puede observar el comportamiento correcto o incorrecto del sistema de control permitiendo, por contraste, reafirmar los aspectos teóricos de la metodología de implementación de sistemas de tiempo real.

Research paper thumbnail of The Emergency of Nutraceutical Compounds in the Preventive Medicine Scenario. Potential for Treatment of Alzheimer's Disease and Other Chronic Disorders

Journal of Alzheimer’s Disease & Parkinsonism, 2018

Evidence-based Nutraceutical Compounds (EBNC), containing bioactive principles of demonstrated ef... more Evidence-based Nutraceutical Compounds (EBNC), containing bioactive principles of demonstrated efficacy and health security are opening solutions for a modern preventive medicine, and as potential solutions for many chronic diseases of the human beings. EBNC contain bioactive components of the human diet that can be used for the prevention or treatment of a disease. They are obtained through rigorous processes of extraction from natural resources and Good Manufacturing Practices (GMP) regulations, and exhibit sound preclinical studies published in high impact medical journals, and double-blind placebo-controlled clinical trials. Are these compounds significantly more effective than alternative medicines? EBNC rely on some of the major advances in molecular genetics, epitranscriptomics, molecular biology and modern pharmacology. They are certainly opening a solid pathway in benefit of human health and the welfare of mankind.

Research paper thumbnail of Neuroinflammation and Neurodegeneration

Update on Dementia, 2016

Pathophysiological processes of neurodegenerative diseases are not clearly defined. However, an i... more Pathophysiological processes of neurodegenerative diseases are not clearly defined. However, an important body of evidence points toward the role of various inflammatory processes. The microglial cell is the main representative of the immune system in the central nervous system (CNS). This cell type can sense foreign or harmful pathogens and trigger its own activation and the generation of neuroinflammatory processes through phagocytosis and the release of cytokines, in order to maintain the cellular microenvironment. However, after maintaining a permanent state of activation due to sustained stimulation over time, microglial cells may generate a focus of persistent inflammation that in some cases precedes or enhances the neurodegenerative process. Thus, neuroinflammatory microenvironment becomes toxic and harmful for the neuronal cell, which degenerates and releases various factors that in turn activate the inflammatory response of microglia, potentiating the neurodegenerative cycle. In this chapter, we discuss the evidence on the role of microglial cell activation in neurodegenerative conditions and the association between neuroinflammatory processes and agerelated neurological diseases. Finally, we outline how this new approach can help us to find new ways to understand neurodegenerative processes and to orientate the search for new therapies.

Research paper thumbnail of Detection of Human Gait Phases Using Textile Pressure Sensors: A Low Cost and Pervasive Approach

Sensors

Human gait analysis is a standard method used for detecting and diagnosing diseases associated wi... more Human gait analysis is a standard method used for detecting and diagnosing diseases associated with gait disorders. Wearable technologies, due to their low costs and high portability, are increasingly being used in gait and other medical analyses. This paper evaluates the use of low-cost homemade textile pressure sensors to recognize gait phases. Ten sensors were integrated into stretch pants, achieving an inexpensive and pervasive solution. Nevertheless, such a simple fabrication process leads to significant sensitivity variability among sensors, hindering their adoption in precision-demanding medical applications. To tackle this issue, we evaluated the textile sensors for the classification of gait phases over three machine learning algorithms for time-series signals, namely, random forest (RF), time series forest (TSF), and multi-representation sequence learner (Mr-SEQL). Training and testing signals were generated from participants wearing the sensing pants in a test run under l...

Research paper thumbnail of Thermal Face Generation Using StyleGAN

IEEE Access, 2021

This article proposes the use of generative adversarial networks (GANs) via StyleGAN2 to create h... more This article proposes the use of generative adversarial networks (GANs) via StyleGAN2 to create high-quality synthetic thermal images and obtain training data to build thermal face recognition models using deep learning. We employed different variants of StyleGAN2, incorporating the new improved version of StyleGAN that uses adaptive discriminator augmentation (ADA). In addition, three different thermal databases from the literature were employed to train a thermal face detector based on YOLOv3 and to train StyleGAN2 and its variants, evaluating different metrics. The synthetic thermal database was built using GANSpace to manipulate the intermediate latent space w of StyleGAN2 and obtain images with different characteristics, such as eyeglasses, rotation, beards, etc. We carried out the training of 6 pretrained deep learning models for face recognition to validate the use of our synthetic thermal database, obtaining 99.98% accuracy for classifying synthetic thermal face images. INDEX TERMS Generative adversarial networks, StyleGAN2, thermal face recognition, deep learning.

Research paper thumbnail of Position control of a mobile robot using reinforcement learning

Research paper thumbnail of Online Virtual Control Laboratory of Mobile Robots

Research paper thumbnail of Face Recognition and Drunk Classification Using Infrared Face Images

Journal of Sensors, 2018

The aim of this study is to propose a system that is capable of recognising the identity of a per... more The aim of this study is to propose a system that is capable of recognising the identity of a person, indicating whether the person is drunk using only information extracted from thermal face images. The proposed system is divided into two stages, face recognition and classification. In the face recognition stage, test images are recognised using robust face recognition algorithms: Weber local descriptor (WLD) and local binary pattern (LBP). The classification stage uses Fisher linear discriminant to reduce the dimensionality of the features, and those features are classified using a classifier based on a Gaussian mixture model, creating a classification space for each person, extending the state-of-the-art concept of a “DrunkSpace Classifier.” The system was validated using a new drunk person database, which was specially designed for this work. The main results show that the performance of the face recognition stage was 100% with both algorithms, while the drunk identification saw...

Research paper thumbnail of A Neural Network Approach for Building An Obstacle Detection Model by Fusion of Proximity Sensors Data

Sensors (Basel, Switzerland), Jan 25, 2018

Proximity sensors are broadly used in mobile robots for obstacle detection. The traditional calib... more Proximity sensors are broadly used in mobile robots for obstacle detection. The traditional calibration process of this kind of sensor could be a time-consuming task because it is usually done by identification in a manual and repetitive way. The resulting obstacles detection models are usually nonlinear functions that can be different for each proximity sensor attached to the robot. In addition, the model is highly dependent on the type of sensor (e.g., ultrasonic or infrared), on changes in light intensity, and on the properties of the obstacle such as shape, colour, and surface texture, among others. That is why in some situations it could be useful to gather all the measurements provided by different kinds of sensor in order to build a unique model that estimates the distances to the obstacles around the robot. This paper presents a novel approach to get an obstacles detection model based on the fusion of sensors data and automatic calibration by using artificial neural networks.

Research paper thumbnail of Detection of Human Gait Phases Using Textile Pressure Sensors: A Low Cost and Pervasive Approach

Sensors

Human gait analysis is a standard method used for detecting and diagnosing diseases associated wi... more Human gait analysis is a standard method used for detecting and diagnosing diseases associated with gait disorders. Wearable technologies, due to their low costs and high portability, are increasingly being used in gait and other medical analyses. This paper evaluates the use of low-cost homemade textile pressure sensors to recognize gait phases. Ten sensors were integrated into stretch pants, achieving an inexpensive and pervasive solution. Nevertheless, such a simple fabrication process leads to significant sensitivity variability among sensors, hindering their adoption in precision-demanding medical applications. To tackle this issue, we evaluated the textile sensors for the classification of gait phases over three machine learning algorithms for time-series signals, namely, random forest (RF), time series forest (TSF), and multi-representation sequence learner (Mr-SEQL). Training and testing signals were generated from participants wearing the sensing pants in a test run under l...

Research paper thumbnail of Thermal Face Generation Using StyleGAN

IEEE Access, 2021

This article proposes the use of generative adversarial networks (GANs) via StyleGAN2 to create h... more This article proposes the use of generative adversarial networks (GANs) via StyleGAN2 to create high-quality synthetic thermal images and obtain training data to build thermal face recognition models using deep learning. We employed different variants of StyleGAN2, incorporating the new improved version of StyleGAN that uses adaptive discriminator augmentation (ADA). In addition, three different thermal databases from the literature were employed to train a thermal face detector based on YOLOv3 and to train StyleGAN2 and its variants, evaluating different metrics. The synthetic thermal database was built using GANSpace to manipulate the intermediate latent space w of StyleGAN2 and obtain images with different characteristics, such as eyeglasses, rotation, beards, etc. We carried out the training of 6 pretrained deep learning models for face recognition to validate the use of our synthetic thermal database, obtaining 99.98% accuracy for classifying synthetic thermal face images. INDEX TERMS Generative adversarial networks, StyleGAN2, thermal face recognition, deep learning.

Research paper thumbnail of Position control of a mobile robot using reinforcement learning

Research paper thumbnail of Online Virtual Control Laboratory of Mobile Robots

Research paper thumbnail of Face Recognition and Drunk Classification Using Infrared Face Images

Journal of Sensors, 2018

The aim of this study is to propose a system that is capable of recognising the identity of a per... more The aim of this study is to propose a system that is capable of recognising the identity of a person, indicating whether the person is drunk using only information extracted from thermal face images. The proposed system is divided into two stages, face recognition and classification. In the face recognition stage, test images are recognised using robust face recognition algorithms: Weber local descriptor (WLD) and local binary pattern (LBP). The classification stage uses Fisher linear discriminant to reduce the dimensionality of the features, and those features are classified using a classifier based on a Gaussian mixture model, creating a classification space for each person, extending the state-of-the-art concept of a “DrunkSpace Classifier.” The system was validated using a new drunk person database, which was specially designed for this work. The main results show that the performance of the face recognition stage was 100% with both algorithms, while the drunk identification saw...

Research paper thumbnail of Study of Visible Face Recognition Methods Applied to Infrared Spectrum

6th Chilean Conference on Pattern Recognition (CCPR), 2014

Research paper thumbnail of A Neural Network Approach for Building An Obstacle Detection Model by Fusion of Proximity Sensors Data

Sensors (Basel, Switzerland), Jan 25, 2018

Proximity sensors are broadly used in mobile robots for obstacle detection. The traditional calib... more Proximity sensors are broadly used in mobile robots for obstacle detection. The traditional calibration process of this kind of sensor could be a time-consuming task because it is usually done by identification in a manual and repetitive way. The resulting obstacles detection models are usually nonlinear functions that can be different for each proximity sensor attached to the robot. In addition, the model is highly dependent on the type of sensor (e.g., ultrasonic or infrared), on changes in light intensity, and on the properties of the obstacle such as shape, colour, and surface texture, among others. That is why in some situations it could be useful to gather all the measurements provided by different kinds of sensor in order to build a unique model that estimates the distances to the obstacles around the robot. This paper presents a novel approach to get an obstacles detection model based on the fusion of sensors data and automatic calibration by using artificial neural networks.

Research paper thumbnail of Study of Visible Face Recognition Methods Applied to Infrared Spectrum

6th Chilean Conference on Pattern Recognition (CCPR), 2014

Research paper thumbnail of Thermal face recognition over time using sparse representation approach

6th Chilean Conference on Pattern Recognition (CCPR), 2014

Research paper thumbnail of Thermal face recognition over time using sparse representation approach

6th Chilean Conference on Pattern Recognition (CCPR), 2014

Research paper thumbnail of Fusion of Visible and Thermal Descriptors Using Genetic Algorithms for Face Recognition Systems

Research paper thumbnail of Laboratorio de prácticas para la enseñanza de sistemas de control de tiempo real

Resumen: El presente trabajo describe el diseño y desarrollo de un laboratorio de prácticas espec... more Resumen: El presente trabajo describe el diseño y desarrollo de un laboratorio de prácticas especialmente concebido para apoyar el aprendizaje en sistemas de control de tiempo real. La herramienta desarrollada permite realizar experiencias de control de tiempo real sobre un motor de corriente continua tanto en modo simulación (basado en el modelo del proceso) o bien realizando pruebas prácticas usando el motor físico. Así, mediante la asignación de periodos de muestreo, tiempos de cómputo de tareas y prioridades de ejecución, el usuario final de la aplicación (profesores y estudiantes) puede observar el comportamiento correcto o incorrecto del sistema de control permitiendo, por contraste, reafirmar los aspectos teóricos de la metodología de implementación de sistemas de tiempo real.

Research paper thumbnail of The Emergency of Nutraceutical Compounds in the Preventive Medicine Scenario. Potential for Treatment of Alzheimer's Disease and Other Chronic Disorders

Journal of Alzheimer’s Disease & Parkinsonism, 2018

Evidence-based Nutraceutical Compounds (EBNC), containing bioactive principles of demonstrated ef... more Evidence-based Nutraceutical Compounds (EBNC), containing bioactive principles of demonstrated efficacy and health security are opening solutions for a modern preventive medicine, and as potential solutions for many chronic diseases of the human beings. EBNC contain bioactive components of the human diet that can be used for the prevention or treatment of a disease. They are obtained through rigorous processes of extraction from natural resources and Good Manufacturing Practices (GMP) regulations, and exhibit sound preclinical studies published in high impact medical journals, and double-blind placebo-controlled clinical trials. Are these compounds significantly more effective than alternative medicines? EBNC rely on some of the major advances in molecular genetics, epitranscriptomics, molecular biology and modern pharmacology. They are certainly opening a solid pathway in benefit of human health and the welfare of mankind.

Research paper thumbnail of Neuroinflammation and Neurodegeneration

Update on Dementia, 2016

Pathophysiological processes of neurodegenerative diseases are not clearly defined. However, an i... more Pathophysiological processes of neurodegenerative diseases are not clearly defined. However, an important body of evidence points toward the role of various inflammatory processes. The microglial cell is the main representative of the immune system in the central nervous system (CNS). This cell type can sense foreign or harmful pathogens and trigger its own activation and the generation of neuroinflammatory processes through phagocytosis and the release of cytokines, in order to maintain the cellular microenvironment. However, after maintaining a permanent state of activation due to sustained stimulation over time, microglial cells may generate a focus of persistent inflammation that in some cases precedes or enhances the neurodegenerative process. Thus, neuroinflammatory microenvironment becomes toxic and harmful for the neuronal cell, which degenerates and releases various factors that in turn activate the inflammatory response of microglia, potentiating the neurodegenerative cycle. In this chapter, we discuss the evidence on the role of microglial cell activation in neurodegenerative conditions and the association between neuroinflammatory processes and agerelated neurological diseases. Finally, we outline how this new approach can help us to find new ways to understand neurodegenerative processes and to orientate the search for new therapies.