Enrique Cabello - Academia.edu (original) (raw)
Uploads
Papers by Enrique Cabello
Multimedia Tools and Applications, 2019
Neural Computing and Applications, 2015
ABSTRACT
Advances in distributed computing and artificial intelligence journal, Jan 24, 2023
In this paper, a novel methodology for the representation and distance measurement of trajectorie... more In this paper, a novel methodology for the representation and distance measurement of trajectories is introduced in order to perform outliers detection tasks. First, a features extraction procedure based on the linear segmentation of trajectories is presented. Next, a configurable context-aware distance is defined. Our representation and distance are significant in that they weigh the relative importance of several relevant features of the trajectories. A clustering method is applied based on the distances matrix and the outliers detection task is performed in any of the clusters. The results of the experiments show the good performance of the method when applied in two different real data sets.
This paper presents an artificial vision based video-sensor designed to detect pedestrian-vehicle... more This paper presents an artificial vision based video-sensor designed to detect pedestrian-vehicle conflicts at crossing points. This video sensor detects moving objects by isolating them from the background. Then the speed and trajectories are estimated using a Kalman filter. Potential conflicts are then predicted. This system has been tested at two real crossing points at the city of Salamanca (Spain).
Expert Systems with Applications, 2019
Engineering Applications of Artificial Intelligence, 2019
Abstract The analysis of human behavior is a popular topic of research since it allows obtaining ... more Abstract The analysis of human behavior is a popular topic of research since it allows obtaining specific information about individuals, their motivations, and the problems and difficulties they can encounter. Human behavior can be grouped to elaborate profiles that would enable the classification of individuals. Nevertheless, the elaboration of profiles related to human behaviors presents some difficulties associated with the volume of data and the number of parameters typically considered. Thus, the development of software able to automatize the manipulation of data through graphical assistants to produce understandable visualizations of the human behaviors is crucial. In this paper, the VISUVER framework is presented. It uses finite state machines to represent and visualize the dynamic human behavior automatically. This behavior could be provided by real data collected by specific sensors or simulated data. The state machines are built in sequential steps in order to illustrate the dynamic evolution of the behavior over time. VISUVER also includes similarity metrics based on text mining techniques to establish possible profiles among the analyzed behaviors. The Intelligent Transportation Systems (ITS) domain has been considered in order to validate the proposal.
ABSTRACT In this paper, we describe a new framework to detect relevant driving risk variables. Th... more ABSTRACT In this paper, we describe a new framework to detect relevant driving risk variables. The methodology is applied over data sets obtained from a high immersive cabin truck simulator in natural driving conditions. A nonparametric model, based in Nearest Neighbors combined with Restricted Least Squared methods is developed. Three experts were asked to evaluate the driving risk in a truck simulator where the vehicle dynamics factors were stored. Moreover the methodology is extended in order to show the characteristics that they have implicitly selected as the relevant ones.
EURASIP Journal on Image and Video Processing, 2015
Lecture Notes in Computational Vision and Biomechanics, 2014
This chapter presents a novel biologically-inspired system capable of detecting and tracking the ... more This chapter presents a novel biologically-inspired system capable of detecting and tracking the hands of the driver during the driving task. The system needs neither marks nor special device in the hands, so a totally natural driving is allowed. Visual information acquisition has been made using an innovative dynamic vision sensor (DVS) that discards the frame concept entirely, encoding the information in the form of Address-Event-Representation (AER) data. This representation allows the information transmission and processing at the same time. An algorithm for detecting and tracking hands using this information in real time is presented. This method has been designed to work with infra-red visual information, avoiding the dependence of the illumination conditions. The proposed system has been integrated in a highly realistic car simulator and several tests have been carried out. Detailed experiments showing the improvement of using AER representation are presented. The presented work is the first approach to introduce AER technology in an automotive environment.
Proceedings of the 12th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications
We present a face verification system. A 100 people face database has been created using a CCD vi... more We present a face verification system. A 100 people face database has been created using a CCD video camera, with controlled illumination conditions and frontal upright face position. A Principal Component Analysis matrix has been computed with eight images per person, and only the 150 most important eigenvalues have been used. The results of PCA are fed into two classifiers (SVM and RBF), in order to perform a verification process in a control access. The algorithm proposed here allows to compute automatically the optimal acceptance threshold to divide a population of candidates into genuine or registered and impostors or non-registered. A Bayesian approach based on screening techniques has been considered, so that the user provides the economical cost for false acceptances and false rejections within the system. According to the ratio between these two costs, the optimal acceptance threshold is computed as the value that minimizes the expected total cost for both acceptances and r...
Teoría de la Educación: Educación y Cultura en la Sociedad de la Información, 2006
In this paper we present a summary of the main techniques for the automatic recognition of human ... more In this paper we present a summary of the main techniques for the automatic recognition of human emotions. We study the most important kind of emotions, centering on the specific emotions. We focus on the two most relevant ways in which emotions are studied: facial expressions from a video, and phonetic expressions from a speech. Since the type of emotions obtained from these two approaches could be different, the fusion of these different sources of information is one of the most interesting areas of research. We present a review of the statistical techniques used in emotion recognition. A fundamental task is the development of an appropriate database, with real audio-visual data, focus on emotion learning.
Multimedia Tools and Applications, 2019
Neural Computing and Applications, 2015
ABSTRACT
Advances in distributed computing and artificial intelligence journal, Jan 24, 2023
In this paper, a novel methodology for the representation and distance measurement of trajectorie... more In this paper, a novel methodology for the representation and distance measurement of trajectories is introduced in order to perform outliers detection tasks. First, a features extraction procedure based on the linear segmentation of trajectories is presented. Next, a configurable context-aware distance is defined. Our representation and distance are significant in that they weigh the relative importance of several relevant features of the trajectories. A clustering method is applied based on the distances matrix and the outliers detection task is performed in any of the clusters. The results of the experiments show the good performance of the method when applied in two different real data sets.
This paper presents an artificial vision based video-sensor designed to detect pedestrian-vehicle... more This paper presents an artificial vision based video-sensor designed to detect pedestrian-vehicle conflicts at crossing points. This video sensor detects moving objects by isolating them from the background. Then the speed and trajectories are estimated using a Kalman filter. Potential conflicts are then predicted. This system has been tested at two real crossing points at the city of Salamanca (Spain).
Expert Systems with Applications, 2019
Engineering Applications of Artificial Intelligence, 2019
Abstract The analysis of human behavior is a popular topic of research since it allows obtaining ... more Abstract The analysis of human behavior is a popular topic of research since it allows obtaining specific information about individuals, their motivations, and the problems and difficulties they can encounter. Human behavior can be grouped to elaborate profiles that would enable the classification of individuals. Nevertheless, the elaboration of profiles related to human behaviors presents some difficulties associated with the volume of data and the number of parameters typically considered. Thus, the development of software able to automatize the manipulation of data through graphical assistants to produce understandable visualizations of the human behaviors is crucial. In this paper, the VISUVER framework is presented. It uses finite state machines to represent and visualize the dynamic human behavior automatically. This behavior could be provided by real data collected by specific sensors or simulated data. The state machines are built in sequential steps in order to illustrate the dynamic evolution of the behavior over time. VISUVER also includes similarity metrics based on text mining techniques to establish possible profiles among the analyzed behaviors. The Intelligent Transportation Systems (ITS) domain has been considered in order to validate the proposal.
ABSTRACT In this paper, we describe a new framework to detect relevant driving risk variables. Th... more ABSTRACT In this paper, we describe a new framework to detect relevant driving risk variables. The methodology is applied over data sets obtained from a high immersive cabin truck simulator in natural driving conditions. A nonparametric model, based in Nearest Neighbors combined with Restricted Least Squared methods is developed. Three experts were asked to evaluate the driving risk in a truck simulator where the vehicle dynamics factors were stored. Moreover the methodology is extended in order to show the characteristics that they have implicitly selected as the relevant ones.
EURASIP Journal on Image and Video Processing, 2015
Lecture Notes in Computational Vision and Biomechanics, 2014
This chapter presents a novel biologically-inspired system capable of detecting and tracking the ... more This chapter presents a novel biologically-inspired system capable of detecting and tracking the hands of the driver during the driving task. The system needs neither marks nor special device in the hands, so a totally natural driving is allowed. Visual information acquisition has been made using an innovative dynamic vision sensor (DVS) that discards the frame concept entirely, encoding the information in the form of Address-Event-Representation (AER) data. This representation allows the information transmission and processing at the same time. An algorithm for detecting and tracking hands using this information in real time is presented. This method has been designed to work with infra-red visual information, avoiding the dependence of the illumination conditions. The proposed system has been integrated in a highly realistic car simulator and several tests have been carried out. Detailed experiments showing the improvement of using AER representation are presented. The presented work is the first approach to introduce AER technology in an automotive environment.
Proceedings of the 12th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications
We present a face verification system. A 100 people face database has been created using a CCD vi... more We present a face verification system. A 100 people face database has been created using a CCD video camera, with controlled illumination conditions and frontal upright face position. A Principal Component Analysis matrix has been computed with eight images per person, and only the 150 most important eigenvalues have been used. The results of PCA are fed into two classifiers (SVM and RBF), in order to perform a verification process in a control access. The algorithm proposed here allows to compute automatically the optimal acceptance threshold to divide a population of candidates into genuine or registered and impostors or non-registered. A Bayesian approach based on screening techniques has been considered, so that the user provides the economical cost for false acceptances and false rejections within the system. According to the ratio between these two costs, the optimal acceptance threshold is computed as the value that minimizes the expected total cost for both acceptances and r...
Teoría de la Educación: Educación y Cultura en la Sociedad de la Información, 2006
In this paper we present a summary of the main techniques for the automatic recognition of human ... more In this paper we present a summary of the main techniques for the automatic recognition of human emotions. We study the most important kind of emotions, centering on the specific emotions. We focus on the two most relevant ways in which emotions are studied: facial expressions from a video, and phonetic expressions from a speech. Since the type of emotions obtained from these two approaches could be different, the fusion of these different sources of information is one of the most interesting areas of research. We present a review of the statistical techniques used in emotion recognition. A fundamental task is the development of an appropriate database, with real audio-visual data, focus on emotion learning.