DI - GIAA - Artículos de Revistas

Permanent URI for this collection

http://dspace.uc3m.es/handle/10016/7454

Browse

Now showing 1 - 20 of 138

3D Object Detection for Self-Driving Cars Using Video and LiDAR: An Ablation Study
(MDPI, 2023-03-01) Salmane, Pascal Housam; Rivera Velázquez, Josué Manuel; Khoudour, Louahdi; Mai, Nguyen Anh Minh; Duthon, Pierre; Crouzil, Alain; Saint Pierre, Guillaume; Velastin Carroza, Sergio Alejandro
Methods based on 64-beam LiDAR can provide very precise 3D object detection. However, highly accurate LiDAR sensors are extremely costly: a 64-beam model can cost approximately USD 75,000. We previously proposed SLS–Fusion (sparse LiDAR and stereo fusion) to fuse low-cost fourbeam LiDAR with stereo cameras that outperform most advanced stereo–LiDAR fusion methods. In this paper, and according to the number of LiDAR beams used, we analyzed how the stereo and LiDAR sensors contributed to the performance of the SLS–Fusion model for 3D object detection. Data coming from the stereo camera play a significant role in the fusion model. However, it is necessary to quantify this contribution and identify the variations in such a contribution with respect to the number of LiDAR beams used inside the model. Thus, to evaluate the roles of the parts of the SLS–Fusion network that represent LiDAR and stereo camera architectures, we propose dividing the model into two independent decoder networks. The results of this study show that—starting from four beams—increasing the number of LiDAR beams has no significant impact on the SLS–Fusion performance. The presented results can guide the design decisions by practitioners.
LIPSNN: A Light Intrusion-Proving Siamese Neural Network Model for Facial Verification
(UNIR La Universidad en Internet, 2022-06-01) Alcaide, Asier; Patricio Guisado, Miguel Ángel; Berlanga de Jesús, Antonio; Arroyo, Angel; Cuadrado Gallego, Juan José; Comunidad de Madrid; Ministerio de Economía y Competitividad (España); Universidad Carlos III de Madrid
Facial verification has experienced a breakthrough in recent years, not only due to the improvement in accuracy of the verification systems but also because of their increased use. One of the main reasons for this has been the appearance and use of new models of Deep Learning to address this problem. This extension in the use of facial verification has had a high impact due to the importance of its applications, especially on security, but the extension of its use could be significantly higher if the problem of the required complex calculations needed by the Deep Learning models, that usually need to be executed on machines with specialised hardware, were solved. That would allow the use of facial verification to be extended, making it possible to run this software on computers with low computing resources, such as Smartphones or tablets. To solve this problem, this paper presents the proposal of a new neural model, called Light Intrusion-Proving Siamese Neural Network, LIPSNN. This new light model, which is based on Siamese Neural Networks, is fully presented from the description of its two block architecture, going through its development, including its training with the well- known dataset Labeled Faces in the Wild, LFW; to its benchmarking with other traditional and deep learning models for facial verification in order to compare its performance for its use in low computing resources systems for facial recognition. For this comparison the attribute parameters, storage, accuracy and precision have been used, and from the results obtained it can be concluded that the LIPSNN can be an alternative to the existing models to solve the facet problem of running facial verification in low computing resource devices.
Data Association Methodology to Improve Spatial Predictions in Alternative Marketing Circuits in Ecuador
(2018-11-05) Padilla, Washington R.; García Herrero, Jesús; Ministerio de Economía y Competitividad (España)
This work proposes a methodology that reduces the error of future estimations in commercialization based on multivariate spatial prediction techniques (cokriging) considering the products with strong associations. It is based on the Apriori algorithm to find association rules in sales of agricultural products of local markets. Results show the improvement in spatial prediction accuracy after using the best association rules.
State Estimation Fusion for Linear Microgrids over an Unreliable Network
(MDPI, 2022-03-21) Soleymannejad, Mohammad; Sadrian Zadeh, Danial; Moshiri, Behzad; Sadjadi, Ebrahim; García Herrero, Jesús; Molina López, José Manuel; Comunidad de Madrid; Ministerio de Ciencia e Innovación (España); Universidad Carlos III de Madrid
Microgrids should be continuously monitored in order to maintain suitable voltages over time. Microgrids are mainly monitored remotely, and their measurement data transmitted through lossy communication networks are vulnerable to cyberattacks and packet loss. The current study leverages the idea of data fusion to address this problem. Hence, this paper investigates the effects of estimation fusion using various machine-learning (ML) regression methods as data fusion methods by aggregating the distributed Kalman filter (KF)-based state estimates of a linear smart microgrid in order to achieve more accurate and reliable state estimates. This unreliability in measurements is because they are received through a lossy communication network that incorporates packet loss and cyberattacks. In addition to ML regression methods, multi-layer perceptron (MLP) and dependent ordered weighted averaging (DOWA) operators are also employed for further comparisons. The results of simulation on the IEEE 4-bus model validate the effectiveness of the employed ML regression methods through the RMSE, MAE and R-squared indices under the condition of missing and manipulated measurements. In general, the results obtained by the Random Forest regression method were more accurate than those of other methods.
A neural network approach to intention modeling for user-adapted conversational agents
(Hindawi, 2016-12) Griol Barres, David; Callejas, Zoraida
Spoken dialogue systems have been proposed to enable a more natural and intuitive interaction with the environment and human-computer interfaces. In this contribution, we present a framework based on neural networks that allows modeling of the user's intention during the dialogue and uses this prediction to dynamically adapt the dialogue model of the system taking into consideration the user's needs and preferences. We have evaluated our proposal to develop a user-adapted spoken dialogue system that facilitates tourist information and services and provide a detailed discussion of the positive influence of our proposal in the success of the interaction, the information and services provided, and the quality perceived by the users.
CONEqNet: convolutional music equalizer network
(Springer Nature, 2023-01-01) Iriz González, Jesús; Patricio Guisado, Miguel Ángel; Berlanga de Jesús, Antonio; Molina López, José Manuel; Comunidad de Madrid; Ministerio de Ciencia e Innovación (España)
The process of parametric equalization of musical pieces seeks to highlight their qualities by cutting and/or stimulating certain frequencies. In this work, we present a neural model capable of equalizing a song according to the musical genre that is being played at a given moment. It is normal that (1) the equalization should adapt throughout the song and not always be the same for the whole song; and (2) songs do not always belong to a specific musical genre and may contain touches of different musical genres. The neural model designed in this work, called CONEqNet (convolutional music equalizer network), takes these aspects into account and proposes a neural model capable of adapting to the different changes that occur throughout a song and with the possibility of mixing nuances of different musical genres. For the training of this model, the well-known GTzan dataset, which provides 1,000 fragments of songs of 30 seconds each, divided into 10 genres, was used. The paper will show proofs of concept of the performance of the neural model.
Error Reduction in Vision-Based Multirotor Landing System
(MDPI, 2022-05-10) Llerena Caña, Juan Pedro; García Herrero, Jesús; Molina López, José Manuel; Comunidad de Madrid; Ministerio de Ciencia e Innovación (España); Universidad Carlos III de Madrid
New applications are continuously appearing with drones as protagonists, but all of them share an essential critical maneuver—landing. New application requirements have led the study of novel landing strategies, in which vision systems have played and continue to play a key role. Generally, the new applications use the control and navigation systems embedded in the aircraft. However, the internal dynamics of these systems, initially focused on other tasks such as the smoothing trajectories between different waypoints, can trigger undesired behaviors. In this paper, we propose a landing system based on monocular vision and navigation information to estimate the helipad global position. In addition, the global estimation system includes a position error correction module by cylinder space transformation and a filtering system with a sliding window. To conclude, the landing system is evaluated with three quality metrics, showing how the proposed correction system together with stationary filtering improves the raw landing system.
Review and classification of trajectory summarisation algorithms: From compression to segmentation
(Sage Journals, 2021-10-30) Amigo Herrero, Daniel; Sánchez Pedroche, David; García Herrero, Jesús; Molina López, José Manuel; Ministerio de Economía y Competitividad (España)
With the continuous development and cost reduction of positioning and tracking technologies, a large amount of trajectories are being exploited in multiple domains for knowledge extraction. A trajectory is formed by a large number of measurements, where many of them are unnecessary to describe the actual trajectory of the vehicle, or even harmful due to sensor noise. This not only consumes large amounts of memory, but also makes the extracting knowledge process more difficult. Trajectory summarisation techniques can solve this problem, generating a smaller and more manageable representation and even semantic segments. In this comprehensive review, we explain and classify techniques for the summarisation of trajectories according to their search strategy and point evaluation criteria, describing connections with the line simplification problem. We also explain several special concepts in trajectory summarisation problem. Finally, we outline the recent trends and best practices to continue the research in next summarisation algorithms.
BERT for Activity Recognition Using Sequences of Skeleton Features and Data Augmentation with GAN
(MDPI, 2023-02-01) Ramirez, Heilym; Velastin Carroza, Sergio Alejandro; Cuéllar, Sara; Fabregas, Ernesto; Farias, Gonzalo; Ministerio de Ciencia e Innovación (España)
Recently, the scientific community has placed great emphasis on the recognition of human activity, especially in the area of health and care for the elderly. There are already practical applications of activity recognition and unusual conditions that use body sensors such as wrist-worn devices or neck pendants. These relatively simple devices may be prone to errors, might be uncomfortable to wear, might be forgotten or not worn, and are unable to detect more subtle conditions such as incorrect postures. Therefore, other proposed methods are based on the use of images and videos to carry out human activity recognition, even in open spaces and with multiple people. However, the resulting increase in the size and complexity involved when using image data requires the use of the most recent advanced machine learning and deep learning techniques. This paper presents an approach based on deep learning with attention to the recognition of activities from multiple frames. Feature extraction is performed by estimating the pose of the human skeleton, and classification is performed using a neural network based on Bidirectional Encoder Representation of Transformers (BERT). This algorithm was trained with the UP-Fall public dataset, generating more balanced artificial data with a Generative Adversarial Neural network (GAN), and evaluated with real data, outperforming the results of other activity recognition methods using the same dataset.
Application of smooth fuzzy model in image denoising and edge detection
(MDPI, 2022-07-02) Sadjadi, Ebrahim; Sadrian Zadeh, Danial; Moshiri, Behzad; García Herrero, Jesús; Molina López, José Manuel; Fernández, Roemi; Comunidad de Madrid; Ministerio de Ciencia e Innovación (España)
In this paper, the bounded variation property of fuzzy models with smooth compositions have been studied, and they have been compared with the standard fuzzy composition (e.g., min-max). Moreover, the contribution of the bounded variation of the smooth fuzzy model for the noise removal and edge preservation of the digital images has been investigated. Different simulations on the test images have been employed to verify the results. The performance index related to the detected edges of the smooth fuzzy models in the presence of both Gaussian and Impulse (also known as salt-and-pepper noise) noises of different densities has been found to be higher than the standard well-known fuzzy models (e.g., min-max composition), which demonstrates the efficiency of smooth compositions in comparison to the standard composition.
State Estimation Fusion for Linear Microgrids over an Unreliable Network
(MDPI AG, 2022-03-21) Soleymannejad, Mohammad; Sadrian Zadeh, Danial; Moshiri, Behzad; Sadjadi, Ebrahim; García Herrero, Jesús; Molina López, José Manuel; Comunidad de Madrid; Ministerio de Ciencia, Innovación y Universidades (España); Universidad Carlos III de Madrid
Microgrids should be continuously monitored in order to maintain suitable voltages over time. Microgrids are mainly monitored remotely, and their measurement data transmitted through lossy communication networks are vulnerable to cyberattacks and packet loss. The current study leverages the idea of data fusion to address this problem. Hence, this paper investigates the effects of estimation fusion using various machine-learning (ML) regression methods as data fusion methods by aggregating the distributed Kalman filter (KF)-based state estimates of a linear smart microgrid in order to achieve more accurate and reliable state estimates. This unreliability in measurements is because they are received through a lossy communication network that incorporates packet loss and cyberattacks. In addition to ML regression methods, multi-layer perceptron (MLP) and dependent ordered weighted averaging (DOWA) operators are also employed for further comparisons. The results of simulation on the IEEE 4-bus model validate the effectiveness of the employed ML regression methods through the RMSE, MAE and R-squared indices under the condition of missing and manipulated measurements. In general, the results obtained by the Random Forest regression method were more accurate than those of other methods.
Error Reduction in Vision-Based Multirotor Landing System
(MDPI AG, 2022-05-10) Llerena Caña, Juan Pedro; García Herrero, Jesús; Molina López, José Manuel; Comunidad de Madrid; Ministerio de Ciencia e Innovación (España); Universidad Carlos III de Madrid
New applications are continuously appearing with drones as protagonists, but all of them share an essential critical maneuver—landing. New application requirements have led the study of novel landing strategies, in which vision systems have played and continue to play a key role. Generally, the new applications use the control and navigation systems embedded in the aircraft. However, the internal dynamics of these systems, initially focused on other tasks such as the smoothing trajectories between different waypoints, can trigger undesired behaviors. In this paper, we propose a landing system based on monocular vision and navigation information to estimate the helipad global position. In addition, the global estimation system includes a position error correction module by cylinder space transformation and a filtering system with a sliding window. To conclude, the landing system is evaluated with three quality metrics, showing how the proposed correction system together with stationary filtering improves the raw landing system.
Human Activity Recognition by Sequences of Skeleton Features
(MDPI AG, 2021-06) Ramirez, Heilym; Velastin Carroza, Sergio Alejandro; Aguayo, Paulo; Fabregas, Ernesto; Farias, Gonzalo; Ministerio de Ciencia e Innovación (España)
In recent years, much effort has been devoted to the development of applications capable of detecting different types of human activity. In this field, fall detection is particularly relevant, especially for the elderly. On the one hand, some applications use wearable sensors that are integrated into cell phones, necklaces or smart bracelets to detect sudden movements of the person wearing the device. The main drawback of these types of systems is that these devices must be placed on a person’s body. This is a major drawback because they can be uncomfortable, in addition to the fact that these systems cannot be implemented in open spaces and with unfamiliar people. In contrast, other approaches perform activity recognition from video camera images, which have many advantages over the previous ones since the user is not required to wear the sensors. As a result, these applications can be implemented in open spaces and with unknown people. This paper presents a vision-based algorithm for activity recognition. The main contribution of this work is to use human skeleton pose estimation as a feature extraction method for activity detection in video camera images. The use of this method allows the detection of multiple people’s activities in the same scene. The algorithm is also capable of classifying multi-frame activities, precisely for those that need more than one frame to be detected. The method is evaluated with the public UP-FALL dataset and compared to similar algorithms using the same dataset.
Variational autoencoders for anomaly detection in the behaviour of the elderly using electricity consumption data
(Wiley Publishing Ltd, 2021-06-15) González, Daniel; Patricio Guisado, Miguel Ángel; Berlanga de Jesús, Antonio; Molina López, José Manuel; Comunidad de Madrid; Ministerio de Economía y Competitividad (España); Universidad Carlos III de Madrid
According To The World Health Organization, Between 2000 And 2050, The Propor Tion Of The World' S Population Over 60 Will Double, From 11% To 22%. In Absolute Numbers, This Age Group Will Increase From 605 Million To 2 Billion In The Course Of Half A Century. It Is A Reality That Most Of Them Prefer To Live Alone, So It Is Necessary To Look For Mechanisms And Tools That Will Help Them To Improve Their Autonomy. Although In Recent Years, We Have Been Living In A Veritable Explosion Of Domotic Sys Tems That Facilitate People' S Daily Lives, It Is Also True That There Are Not Many Tools Specifically Aimed At This Sector Of The Population. The Aim Of This Paper Is To Present A Potential Solution To The Monitoring Of Activity Of Daily Living In The Least Intrusive Way For People. In This Case, Anomalous Patterns Of Daily Activities Will Be Detected By Analysing The Daily Consumption Of Household Appliances. People Who Live Alone Usu Ally Have A Pattern Of Daily Behaviour In The Use Of Household Appliances (Coffee Machine, Microwave, Television, Etc.). A Neuronal Model Is Proposed For The Detection Of Abnormal Behaviour Based On An Autoencoder Architecture. This Solution Will Be Compared With A Variational Autoencoder To Analyse The Improvements That Can Be Obtained. The Well-Known Dataset Called Uk-Dale Will Be Used To Validate The Proposal.
Analysis of scientific production based on trending research topics. An Artificial Intelligence case study
(CSIC, 2019-03-30) Bobadilla, Jesús; Gutiérrez, Abraham; Patricio Guisado, Miguel Ángel; Bojorque, Rodolfo Xavier
Scientific Documentation Research Leads To The Computation Of Large Amounts Of Information From Published Works Of The Scientific Community. It Is Necessary To Explain These Processes And Create Application Frameworks. This Paper Provides The Following: A) An Information System Designed To Extract Scientific Information From Published Papers, B) Accurate Explanations Of The Main Processing Stages Including Data Mining, Natural Language Processing, And Machine Learning, And C) Categorized And Explained Results Coming From The Artificial Intelligence Case Study. The Results In This Paper Include The Following: A) Topics And Research Area Rankings And B) Quantity Versus Quality Comparisons Of Topics And Research Areas.
Bridging from syntactic to statistical methods: Classification with automatically segmented features from sequences
(Elsevier, 2015-11-01) Sidorova, Julia; García Herrero, Jesús
To Integrate The Benefits Of Statistical Methods Into Syntactic Pattern Recognition, A Bridging Approach Is Proposed: (I) Acquisition Of A Grammar Per Recognition Class (Ii) Comparison Of The Obtained Grammars In Order To Find Substructures Of Interest Represented As Sequences Of Terminal And/Or Non-Terminal Symbols And Filling The Feature Vector With Their Counts (Iii) Hierarchical Feature Selection And Hierarchical Classification, Deducing And Accounting For The Domain Taxonomy. The Bridging Approach Has The Benefits Of Syntactic Methods: Preserves Structural Relations And Gives Insights Into The Problem. Yet, It Does Not Imply Distance Calculations And, Thus, Saves A Non-Trivial Task-Dependent Design Step. Instead It Relies On Statistical Classification From Many Features. Our Experiments Concern A Difficult Problem Of Chemical Toxicity Prediction. The Code And The Data Set Are Open-Source. (C) 2015 Elsevier Ltd. All Rights Reserved.
A data-driven approach to spoken dialog segmentation
(Elsevier, 2020-05-28) Griol Barres, David; Molina López, José Manuel; Sanchis de Miguel, María Araceli; Callejas Carrión, Zoraida
In This Paper, We Present A Statistical Model For Spoken Dialog Segmentation That Decides The Current Phase Of The Dialog By Means Of An Automatic Classification Process. We Have Applied Our Proposal To Three Practical Conversational Systems Acting In Different Domains. The Results Of The Evaluation Show That Is Possible To Attain High Accuracy Rates In Dialog Segmentation When Using Different Sources Of Information To Represent The User Input. Our Results Indicate How The Module Proposed Can Also Improve Dialog Management By Selecting Better System Answers. The Statistical Model Developed With Human-Machine Dialog Corpora Has Been Applied In One Of Our Experiments To Human-Human Conversations And Provides A Good Baseline As Well As Insights In The Model Limitation.
A multimodal conversational coach for active ageing based on sentient computing and m-health
(John Wiley & Sons, Ltd, 2020-04-01) Griol Barres, David; Molina López, José Manuel; Sanchis de Miguel, María Araceli; Ministerio de Economía y Competitividad (España)
As Life Expectancy Increases, It Has Become More Necessary To Find Ways To Support Healthy Ageing. A Number Of Active Ageing Initiatives Are Being Developed Nowadays To Foster Healthy Habits In The Population. This Paper Presents Our Contribution To These Initiatives In The Form Of A Multimodal Conversational Coach That Acts As A Coach For Physical Activities. The Agent Can Be Developed As An Android App Running On Smartphones And Coupled With Cheap Widely Available Sport Sensors In Order To Provide Meaningful Coaching. It Can Be Employed To Prepare Exercise Sessions, Provide Feedback During The Sessions, And Discuss The Results After The Exercise. It Incorporates An Affective Component That Informs Dynamic User Models To Produce Adaptive Interaction Strategies.
Fuzzy model identification and self learning with smooth compositions
(Springer, 2019-10-03) Sadjadi, Ebrahim; García Herrero, Jesús; Molina López, José Manuel; Borzabadi, Akbar Hashemi; Abchouyeh, Monireh Asadi; Ministerio de Economía y Competitividad (España)
This Paper Develops A Smooth Model Identification And Self-Learning Strategy For Dynamic Systems Taking Into Account Possible Parameter Variations And Uncertainties. We Have Tried To Solve The Problem Such That The Model Follows The Changes And Variations In The System On A Continuous And Smooth Surface. Running The Model To Adaptively Gain The Optimum Values Of The Parameters On A Smooth Surface Would Facilitate Further Improvements In The Application Of Other Derivative Based Optimization Control Algorithms Such As Mpc Or Robust Control Algorithms To Achieve A Combined Modeling-Control Scheme. Compared To The Earlier Works On The Smooth Fuzzy Modeling Structures, We Could Reach A Desired Trade-Off Between The Model Optimality And The Computational Load. The Proposed Method Has Been Evaluated On A Test Problem As Well As The Non-Linear Dynamic Of A Chemical Process.
MONEDA: scalable multi-objective optimization with a neural network-based estimation of distribution algorithm
(Springer, 2016-12-01) Martí, Luis; García Herrero, Jesús; Berlanga de Jesús, Antonio; Ministerio de Economía y Competitividad (España)
The Extension Of Estimation Of Distribution Algorithms (Edas) To The Multiobjective Domain Has Led To Multi-Objective Optimization Edas (Moedas). Most Moedas Have Limited Themselves To Porting Single-Objective Edas To The Multi-Objective Domain. Although Moedas Have Proved To Be A Valid Approach, The Last Point Is An Obstacle To The Achievement Of A Significant Improvement Regarding "Standard" Multi-Objective Optimization Evolutionary Algorithms. Adapting The Model-Building Algorithm Is One Way To Achieve A Substantial Advance. Most Model-Building Schemes Used So Far By Edas Employ Off-The-Shelf Machine Learning Methods. However, The Model-Building Problem Has Particular Requirements That Those Methods Do Not Meet And Even Evade. The Focus Of This Paper Is On The Model- Building Issue And How It Has Not Been Properly Understood And Addressed By Most Moedas. We Delve Down Into The Roots Of This Matter And Hypothesize About Its Causes. To Gain A Deeper Understanding Of The Subject We Propose A Novel Algorithm Intended To Overcome The Draw-Backs Of Current Moedas. This New Algorithm Is The Multi-Objective Neural Estimation Of Distribution Algorithm (Moneda). Moneda Uses A Modified Growing Neural Gas Network For Model-Building (Mb-Gng). Mb-Gng Is A Custom-Made Clustering Algorithm That Meets The Above Demands. Thanks To Its Custom-Made Model-Building Algorithm, The Preservation Of Elite Individuals And Its Individual Replacement Scheme, Moneda Is Capable Of Scalably Solving Continuous Multi-Objective Optimization Problems. It Performs Better Than Similar Algorithms In Terms Of A Set Of Quality Indicators And Computational Resource Requirements.

Browse

Recent Submissions