Diego Andina | Technical University of Madrid (original) (raw)

Videos by Diego Andina

When Prof. Hojjat Adeli, a Clarivate Analytics Highly Cited Researcher in three categories: Comp... more When Prof. Hojjat Adeli, a Clarivate Analytics Highly Cited
Researcher in three categories: Computer Science, Engineering, and Cross-Field received his Honoris Causa title by Universidad Politécnica de Madrid, surprised me with this generous laudatio.
Thanks Hojjat!.

1996 Documentary for Spanish TV.

Papers by Diego Andina

Springer eBooks, 2007

Page 1. Diego Andina DucTruongPham Editors for Engineering an Page 2. Computational Intelligence ... more Page 1. Diego Andina DucTruongPham Editors for Engineering an Page 2. Computational Intelligence Page 3. Computational Intelligence for Engineering and Manufacturing Edited by Diego Andina Technical University of Madrid ...

World Automation Congress, 2004

The control of the security in a limited area, like a house, is a complex task. This paper propos... more The control of the security in a limited area, like a house, is a complex task. This paper proposc a stand-alone intelligent system based on image recogiiit~on. The system detects moving peisoiis and successive update steps are applied in order to track them pviding important information about the position and their activity in term oi'traiectories perlbrined. Finally, a verification subsystem. based on the Intrusioti Detection System (IDS) used in coinputer networks technology attempts to identify unauthorized proximity based on ntrrinality patterns. 1. IN'I'RC)DUCTION Automatic human detection and body part lucalization arc important and challenging prohleins in compufer vision (1).(2). The solution to those prohleins cilii be einploycd in a wide nnge of applications such as safe robot navigation, visual surveillance, huinan-computer intcrl'ace, and figure animation. Our application doinnin is Domotics, i.e. the integral automation of huildings and housing. We can dcline it as: "A scr of eleincnts that, when installed, interconnected and cotitrolled autoinatically in a building. save the users wonying about routine weiyday actions, providing improvement in their coinfoii. in encrgy consumption. in security and iii conmimicarion as well". The context for human detectivn is the general ob,ject detection problem. Our approach is IO lirst seginetit Ibregruund ob.jects fium the hackground and then classify cach segmented ohject as human or non-human. Classifying only seginentcd ohjeots rather than searching them in the whole image significantly reduces computational complexity. Due to thr application domain of our system, we do not care about the classification of moving objccts as persons. So, we deal every inoving ab,icct as a threat for the security of the house. The tinal Intrusion I3etection subsystem based vti the shape ond trnjectories followed by tlic object should discriminate if this object is compromising the user defined security areas. The orgatiization ol the paper is 21s fAlows. First we review the work done for the European I'ro,ject Eureka EU- 136 I, which makes use of segmentation of forcginund ohjccls to pre-proccss vidcoconference scenes. The dctected ohjects are tracked using a mhiist regression scheme that will dcscribe their trqjectories using affitirr motion inodel parameters. Both the shape and tra.jectorics are used as the input for an Intrusion Detection System ((lF.), which is hascd on previous research done for detecting attacks 011 computer networks (3). The IDS attempts to idcntify unauthorized proximity to the high sccurity areas louking for statistical and rule-bnsed onnmaltes. Thcse high security areas are detined previously by the user of the system with an interact i ve i 11 terfacc.

Springer eBooks, 2011

The training algorithm studied in this paper is inspired by the biological metaplasticity propert... more The training algorithm studied in this paper is inspired by the biological metaplasticity property of neurons. Tested on different multidisciplinary applications, it achieves a more efficient training and improves Artificial Neural Network Performance. The algorithm has been recently proposed for Artificial Neural Networks in general, although for the purpose of discussing its biological plausibility, a Multilayer Perceptron has been used. During the training phase, the artificial metaplasticity multilayer perceptron could be considered a new probabilistic version of the presynaptic rule, as during the training phase the algorithm assigns higher values for updating the weights in the less probable activations than in the ones with higher probability.

World Automation Congress, 2004

One of the mayor limitations associated to Neural Networks (NNs) learning algorithms is the numbe... more One of the mayor limitations associated to Neural Networks (NNs) learning algorithms is the number of input-output pattern pairs needed to obtain the desired classilication pcrlbmancc. In many practical cases, the number 01' pairs used to design the NN is much lower than necessary. Either bccause of the excessivc price of data acquisition or, simply, for availability reasons. An Importance

Lecture notes in electrical engineering, 2010

In the first half of this book, Chapters 1 to 5, we have presented the foundations of information... more In the first half of this book, Chapters 1 to 5, we have presented the foundations of information risk management (Chapter 1), the profiles required by an IT security team (Chapter 2), the basic aspects that guide the team-individual contract (Chapter 3), a list of security principles to follow and activities to perform by the team (Chapter 4) and some

Springer eBooks, Jun 29, 2007

This article presents a neural network model for the simulation of the neurological mechanism tha... more This article presents a neural network model for the simulation of the neurological mechanism that produces limbs hiper-rigidity (spasticity). In this model, we take into account intrinsic plasticity, which is the property of biological neurons that consists in the shifting of the action potential threshold according to experience. In accordance to the computational model, a therapeutic technique for diminishing limbs spasticity is proposed and discussed.

Research on computing science, 2005

International Journal of Pattern Recognition and Artificial Intelligence, Oct 9, 2017

A new sub-segmentation method has been proposed in 2009 which, in digital images, help us to iden... more A new sub-segmentation method has been proposed in 2009 which, in digital images, help us to identify the typical pixels, as well as the less representative pixels or atypical of each segmented region. This method is based on the Possibilistic Fuzzy c-Means (PFCM) clustering algorithm, as it integrates absolute and relative memberships. Now, the segmentation problem is related to isolate each one of the objects present in an image. However, and considering only one segmented object or region represented by gray levels as its only feature, the totality of pixels is divided in two basic groups, the group of pixels representing the object, and the others that do not represent it. In the former group, there is a sub-group of pixels near the most representative element of the object, the prototype, and identified here as the typical pixels, and a sub-group corresponding to the less representative pixels of the object, which are the atypical pixels, and generally located at the borders of the pixels representing the object. Besides, the sub-group of atypical pixels presents greater tones (brighter or towards the white color) or smaller tones (darker or towards black color). So, the sub-segmentation method offers the capability to identify the sub-region of atypical pixels, although without performing a differentiation between the brighter and the darker ones. Hence, the proposal of this work contributes to the problem of image segmentation with the improvement on the detection of the atypical sub-regions, and clearly recognizing between both kind of atypical pixels, because in many cases only the brighter or the darker atypical pixels are the ones that represent the object of interest in an image, depending on the problem to be solved. In this study, two real cases are used to show the contribution of this proposal; the first case serves to demonstrate the pores detection in soil images represented by the darker atypical pixels, and the second one to demonstrate the detection of microcalcifications in mammograms, represented in this case by the brighter atypical pixels.

A Comparison of Criterion Functions for a Neural Network Applied to Binary Detection Diego ANDINA... more A Comparison of Criterion Functions for a Neural Network Applied to Binary Detection Diego ANDINA, Jose L. SANZ-GONZALEZ and Jose.A. JIMENEZ-PAJARES Departamento de Senates, Sistemas y Radiocomunicaciones, ETSI de Telecomunicacion, Universidad ...

Lecture notes in management and industrial engineering, 2020

Springer eBooks, 2017

Rainfall is one of the most important events of human life and society. Some rainfall phenomena l... more Rainfall is one of the most important events of human life and society. Some rainfall phenomena like floods or hailstone are a threat to agriculture, business and even life. Predicting the weather has emerged as one of the most important areas of scientific endeavour. Nowadays, there is a big effort and great developments in long and mid-term rainfall forecasts, where qualitative improvements have been obtained both in forecasts and verification. This work proposes a diverse local rainfall forecasting system, using a long term local measurements registry. The forecast is performed estimating pressure time series and processing them with multispectral wavelet analysis and Neural Networks. The aim of the study is to provide complementary criteria based on the observed pressure wave pattern repetition. This method was proposed by expert meteorologists after observing these events during 40 years.

2018 World Automation Congress (WAC), Jun 1, 2018

Koniocortex-Like Network model is a Bio-Inspired Neural Network structure that tries to replicate... more Koniocortex-Like Network model is a Bio-Inspired Neural Network structure that tries to replicate the architecture and properties of the biological koniocortex section of the brain. The structure is composed by different kinds of artificial neurons that interplay between them to create a competitive model that can be used to classify patterns. The classification performance obtained is based on different properties like lateral inhibition, metaplasticity and intrinsic plasticity, that allows a natural evolution of the network until obtaining the desired results. This kind of network has been applied to synthetic and real data showing big potential, now the network capabilities are tested using other state-of-the-art real data application: the classification of credit data from the Australian Credit Approval Database.

Computación y Sistemas, 2019

In this work, an alternative new methodology to segment regions in an image is proposed. The meth... more In this work, an alternative new methodology to segment regions in an image is proposed. The method takes the advantages offered by hybrid algorithms, as the Possibilistic Fuzzy c-Means (PFCM) clustering algorithm, which has the qualities of both the Fuzzy c-Means (FCM) and the Possibilistic c-Means (PCM). The method is called sub-segmentation, and it consists of finding some clusters in an image through the segmentation of the image and, within these clusters, the less representative pixels or atypical pixels. These elements very frequently represent the zones of interest during image analysis. Three different cases are used in order to illustrate the method. The first one is an image of a drop of milk, where the generality of the method is tested in a simple but representative image. The second case corresponds to digital mammograms, where the potentiality of the method is tested in a critical application, such as anomalies identification in mammograms for cancer detection. The last case gives an idea of its range of applications, as the method is applied to an industrial case of classification of wood boards according to their quality. As can be seen from the three cases used in this work, the results are very interesting and promising.

When Prof. Hojjat Adeli, a Clarivate Analytics Highly Cited Researcher in three categories: Comp... more When Prof. Hojjat Adeli, a Clarivate Analytics Highly Cited
Researcher in three categories: Computer Science, Engineering, and Cross-Field received his Honoris Causa title by Universidad Politécnica de Madrid, surprised me with this generous laudatio.
Thanks Hojjat!.

1996 Documentary for Spanish TV.

Springer eBooks, 2007

Page 1. Diego Andina DucTruongPham Editors for Engineering an Page 2. Computational Intelligence ... more Page 1. Diego Andina DucTruongPham Editors for Engineering an Page 2. Computational Intelligence Page 3. Computational Intelligence for Engineering and Manufacturing Edited by Diego Andina Technical University of Madrid ...

World Automation Congress, 2004

The control of the security in a limited area, like a house, is a complex task. This paper propos... more The control of the security in a limited area, like a house, is a complex task. This paper proposc a stand-alone intelligent system based on image recogiiit~on. The system detects moving peisoiis and successive update steps are applied in order to track them pviding important information about the position and their activity in term oi'traiectories perlbrined. Finally, a verification subsystem. based on the Intrusioti Detection System (IDS) used in coinputer networks technology attempts to identify unauthorized proximity based on ntrrinality patterns. 1. IN'I'RC)DUCTION Automatic human detection and body part lucalization arc important and challenging prohleins in compufer vision (1).(2). The solution to those prohleins cilii be einploycd in a wide nnge of applications such as safe robot navigation, visual surveillance, huinan-computer intcrl'ace, and figure animation. Our application doinnin is Domotics, i.e. the integral automation of huildings and housing. We can dcline it as: "A scr of eleincnts that, when installed, interconnected and cotitrolled autoinatically in a building. save the users wonying about routine weiyday actions, providing improvement in their coinfoii. in encrgy consumption. in security and iii conmimicarion as well". The context for human detectivn is the general ob,ject detection problem. Our approach is IO lirst seginetit Ibregruund ob.jects fium the hackground and then classify cach segmented ohject as human or non-human. Classifying only seginentcd ohjeots rather than searching them in the whole image significantly reduces computational complexity. Due to thr application domain of our system, we do not care about the classification of moving objccts as persons. So, we deal every inoving ab,icct as a threat for the security of the house. The tinal Intrusion I3etection subsystem based vti the shape ond trnjectories followed by tlic object should discriminate if this object is compromising the user defined security areas. The orgatiization ol the paper is 21s fAlows. First we review the work done for the European I'ro,ject Eureka EU- 136 I, which makes use of segmentation of forcginund ohjccls to pre-proccss vidcoconference scenes. The dctected ohjects are tracked using a mhiist regression scheme that will dcscribe their trqjectories using affitirr motion inodel parameters. Both the shape and tra.jectorics are used as the input for an Intrusion Detection System ((lF.), which is hascd on previous research done for detecting attacks 011 computer networks (3). The IDS attempts to idcntify unauthorized proximity to the high sccurity areas louking for statistical and rule-bnsed onnmaltes. Thcse high security areas are detined previously by the user of the system with an interact i ve i 11 terfacc.

Springer eBooks, 2011

The training algorithm studied in this paper is inspired by the biological metaplasticity propert... more The training algorithm studied in this paper is inspired by the biological metaplasticity property of neurons. Tested on different multidisciplinary applications, it achieves a more efficient training and improves Artificial Neural Network Performance. The algorithm has been recently proposed for Artificial Neural Networks in general, although for the purpose of discussing its biological plausibility, a Multilayer Perceptron has been used. During the training phase, the artificial metaplasticity multilayer perceptron could be considered a new probabilistic version of the presynaptic rule, as during the training phase the algorithm assigns higher values for updating the weights in the less probable activations than in the ones with higher probability.

World Automation Congress, 2004

One of the mayor limitations associated to Neural Networks (NNs) learning algorithms is the numbe... more One of the mayor limitations associated to Neural Networks (NNs) learning algorithms is the number of input-output pattern pairs needed to obtain the desired classilication pcrlbmancc. In many practical cases, the number 01' pairs used to design the NN is much lower than necessary. Either bccause of the excessivc price of data acquisition or, simply, for availability reasons. An Importance

Lecture notes in electrical engineering, 2010

In the first half of this book, Chapters 1 to 5, we have presented the foundations of information... more In the first half of this book, Chapters 1 to 5, we have presented the foundations of information risk management (Chapter 1), the profiles required by an IT security team (Chapter 2), the basic aspects that guide the team-individual contract (Chapter 3), a list of security principles to follow and activities to perform by the team (Chapter 4) and some

Springer eBooks, Jun 29, 2007

This article presents a neural network model for the simulation of the neurological mechanism tha... more This article presents a neural network model for the simulation of the neurological mechanism that produces limbs hiper-rigidity (spasticity). In this model, we take into account intrinsic plasticity, which is the property of biological neurons that consists in the shifting of the action potential threshold according to experience. In accordance to the computational model, a therapeutic technique for diminishing limbs spasticity is proposed and discussed.

Research on computing science, 2005

International Journal of Pattern Recognition and Artificial Intelligence, Oct 9, 2017

A new sub-segmentation method has been proposed in 2009 which, in digital images, help us to iden... more A new sub-segmentation method has been proposed in 2009 which, in digital images, help us to identify the typical pixels, as well as the less representative pixels or atypical of each segmented region. This method is based on the Possibilistic Fuzzy c-Means (PFCM) clustering algorithm, as it integrates absolute and relative memberships. Now, the segmentation problem is related to isolate each one of the objects present in an image. However, and considering only one segmented object or region represented by gray levels as its only feature, the totality of pixels is divided in two basic groups, the group of pixels representing the object, and the others that do not represent it. In the former group, there is a sub-group of pixels near the most representative element of the object, the prototype, and identified here as the typical pixels, and a sub-group corresponding to the less representative pixels of the object, which are the atypical pixels, and generally located at the borders of the pixels representing the object. Besides, the sub-group of atypical pixels presents greater tones (brighter or towards the white color) or smaller tones (darker or towards black color). So, the sub-segmentation method offers the capability to identify the sub-region of atypical pixels, although without performing a differentiation between the brighter and the darker ones. Hence, the proposal of this work contributes to the problem of image segmentation with the improvement on the detection of the atypical sub-regions, and clearly recognizing between both kind of atypical pixels, because in many cases only the brighter or the darker atypical pixels are the ones that represent the object of interest in an image, depending on the problem to be solved. In this study, two real cases are used to show the contribution of this proposal; the first case serves to demonstrate the pores detection in soil images represented by the darker atypical pixels, and the second one to demonstrate the detection of microcalcifications in mammograms, represented in this case by the brighter atypical pixels.

A Comparison of Criterion Functions for a Neural Network Applied to Binary Detection Diego ANDINA... more A Comparison of Criterion Functions for a Neural Network Applied to Binary Detection Diego ANDINA, Jose L. SANZ-GONZALEZ and Jose.A. JIMENEZ-PAJARES Departamento de Senates, Sistemas y Radiocomunicaciones, ETSI de Telecomunicacion, Universidad ...

Lecture notes in management and industrial engineering, 2020

Springer eBooks, 2017

Rainfall is one of the most important events of human life and society. Some rainfall phenomena l... more Rainfall is one of the most important events of human life and society. Some rainfall phenomena like floods or hailstone are a threat to agriculture, business and even life. Predicting the weather has emerged as one of the most important areas of scientific endeavour. Nowadays, there is a big effort and great developments in long and mid-term rainfall forecasts, where qualitative improvements have been obtained both in forecasts and verification. This work proposes a diverse local rainfall forecasting system, using a long term local measurements registry. The forecast is performed estimating pressure time series and processing them with multispectral wavelet analysis and Neural Networks. The aim of the study is to provide complementary criteria based on the observed pressure wave pattern repetition. This method was proposed by expert meteorologists after observing these events during 40 years.

2018 World Automation Congress (WAC), Jun 1, 2018

Koniocortex-Like Network model is a Bio-Inspired Neural Network structure that tries to replicate... more Koniocortex-Like Network model is a Bio-Inspired Neural Network structure that tries to replicate the architecture and properties of the biological koniocortex section of the brain. The structure is composed by different kinds of artificial neurons that interplay between them to create a competitive model that can be used to classify patterns. The classification performance obtained is based on different properties like lateral inhibition, metaplasticity and intrinsic plasticity, that allows a natural evolution of the network until obtaining the desired results. This kind of network has been applied to synthetic and real data showing big potential, now the network capabilities are tested using other state-of-the-art real data application: the classification of credit data from the Australian Credit Approval Database.

Computación y Sistemas, 2019

In this work, an alternative new methodology to segment regions in an image is proposed. The meth... more In this work, an alternative new methodology to segment regions in an image is proposed. The method takes the advantages offered by hybrid algorithms, as the Possibilistic Fuzzy c-Means (PFCM) clustering algorithm, which has the qualities of both the Fuzzy c-Means (FCM) and the Possibilistic c-Means (PCM). The method is called sub-segmentation, and it consists of finding some clusters in an image through the segmentation of the image and, within these clusters, the less representative pixels or atypical pixels. These elements very frequently represent the zones of interest during image analysis. Three different cases are used in order to illustrate the method. The first one is an image of a drop of milk, where the generality of the method is tested in a simple but representative image. The second case corresponds to digital mammograms, where the potentiality of the method is tested in a critical application, such as anomalies identification in mammograms for cancer detection. The last case gives an idea of its range of applications, as the method is applied to an industrial case of classification of wood boards according to their quality. As can be seen from the three cases used in this work, the results are very interesting and promising.

Integrated Computer-Aided Engineering, 2018

This paper provides a novel method to design a Rainfall Forecasting System based on Neural Networ... more This paper provides a novel method to design a Rainfall Forecasting System based on Neural Networks and a long-term data registry. The System proposed is based on the observations of expert meteorologists after 40 years of research developing local forecast methodologies to predict rainfall events in the Meteorological Observatory of Valladolid, Spain. The Geostrophic Wind, a theoretical value resulting of the balance between the Pressure Gradient Force and the Coriolis Force, is the key parameter which feeds the aforementioned System. The paper focuses on the Geostrophic Wind calculation. We propose a novel method to estimate its direction and magnitude on a local scale, by processing a set of pressure and temperature observations surrounding a central location. This study continues previous research, providing complementary criteria for numerical weather prediction systems based on time-series forecasting and neural networks.

Neurocomputing, Aug 1, 2013

In this paper, the presynaptic rule, a classical model of synaptic reinforcement, is revisited. I... more In this paper, the presynaptic rule, a classical model of synaptic reinforcement, is revisited. It is shown that this model is capable of reproducing recently discovered properties of biological synapses such as synaptic directionality, and metaplasticity of the long-term potentiation threshold. With slight modifications, the presynaptic model also reproduces metaplasticity of the long-term depression threshold and Artola, Bröcher and Singer's experimental model. Two asymptotically equivalent approaches were adopted for this analysis, one with firing rates and another with conditional probabilities. Although both approximations are consistent with biological properties, the results obtained by the probabilistic approach are qualitatively closer to biological experimental results.

Atypical Data can have high value in Big Data through Machine Learning: https://youtu.be/vYLVNAPHCnI

Emerging and novel Bioinspired Artificial Neural Networks (BIANN) provide new interdisciplinary a... more Emerging and novel Bioinspired Artificial Neural Networks (BIANN) provide new interdisciplinary approaches for solution of complicated and intractable problems. How can engineering, mathematics, computation, Artificial Intelligence (AI) and Knowledge Engineering (KE) find inspiration in the behavior and internal functioning of physical, biological nervous systems to conceive, develop and build-up new concepts, materials, mechanisms and algorithms of potential value for solution of real world applications? And how can these techniques be used to conceptualize and model the nervous system? The aim of this special issue is research on the intersection of neurosciences and artificial neural networks and computations. Novel, emerging, and high impact BIANN theories and models with applications to illustrate their potential are welcome. Please inform the guest editors and the Editor-in-Chief about your intention to submit a manuscript for possible publication in the special issue as soon as possible. Please email the following to one of the Guest Editors with a copy to the Editor-in-Chief by March 15, 2016: