Open Access
Hyperspectral system trade-offs for illumination, hardware and analysis methods: a case study of seed mix ingredient discrimination
(IM Publications, 2020) Blanch Pérez del Notario, Carolina; López Molina, Carlos; Lambrechts, Andy; Saeys, Wouter; Estadística, Informática y Matemáticas; Estatistika, Informatika eta Matematika
The discrimination power of a hyperspectral imaging system for image segmentation or object detection is determined by the illumination, the camera spatial–spectral resolution, and both the pre-processing and analysis methods used for image processing. In this study, we methodically reviewed the alternatives for each of those factors for a case study from the food industry to provide guidance in the construction and configuration of hyperspectral imaging systems in the visible near infrared range for food quality inspection. We investigated both halogen-and LED-based illuminations and considered cameras with different spatial–spectral resolution trade-offs. At the level of the data analysis, we evaluated the impact of binning, median filtering and bilateral filtering as pre-or post-processing and compared pixel-based classifiers with convolutional neural networks for a challenging application in the food industry, namely ingredient identification in a flour–seed mix. Starting from a basic configuration and by modifying the combination of system aspects we were able to increase the mean accuracy by at least 25%. In addition, different trade-offs in performance-complexity were identified for different combinations of system parameters, allowing adaptation to diverse application requirements.
Open Access
Neuro-inspired edge feature fusion using Choquet integrals
(Elsevier, 2021) Marco Detchart, Cedric; Lucca, Giancarlo; López Molina, Carlos; Miguel Turullols, Laura de; Pereira Dimuro, Graçaliz; Bustince Sola, Humberto; Estadística, Informática y Matemáticas; Estatistika, Informatika eta Matematika; Universidad Pública de Navarra / Nafarroako Unibertsitate Publikoa
It is known that the human visual system performs a hierarchical information process in which early vision cues (or primitives) are fused in the visual cortex to compose complex shapes and descriptors. While different aspects of the process have been extensively studied, such as lens adaptation or feature detection, some other aspects, such as feature fusion, have been mostly left aside. In this work, we elaborate on the fusion of early vision primitives using generalizations of the Choquet integral, and novel aggregation operators that have been extensively studied in recent years. We propose to use generalizations of the Choquet integral to sensibly fuse elementary edge cues, in an attempt to model the behaviour of neurons in the early visual cortex. Our proposal leads to a fully-framed edge detection algorithm whose performance is put to the test in state-of-the-art edge detection datasets.
Open Access
Multiscale edge detection using first-order derivative of anisotropic Gaussian kernels
(Springer, 2019) Wang, Gang; López Molina, Carlos; Baets, Bernard de; Estadística, Informática y Matemáticas; Estatistika, Informatika eta Matematika
Spatially scaled edges are ubiquitous in natural images. To better detect edges with heterogeneous widths, in this paper, we propose a multiscale edge detection method based on first-order derivative of anisotropic Gaussian kernels. These kernels are normalized in scale-space, yielding a maximum response at the scale of the observed edge, and accordingly, the edge scale can be identified. Subsequently, the maximum response and the identified edge scale are used to compute the edge strength. Furthermore, we propose an adaptive anisotropy factor of which the value decreases as the kernel scale increases. This factor improves the noise robustness of small-scale kernels while alleviating the anisotropy stretch effect that occurs in conventional anisotropic methods. Finally, we evaluate our method on widely used datasets. Experimental results validate the benefits of our method over the competing methods.
Open Access
A survey of fingerprint classification Part I: taxonomies on feature extraction methods and learning models
(Elsevier, 2015) Galar Idoate, Mikel; Derrac, Joaquín; Peralta, Daniel; Triguero, Isaac; Paternain Dallo, Daniel; López Molina, Carlos; García, Salvador; Benítez, José Manuel; Pagola Barrio, Miguel; Barrenechea Tartas, Edurne; Bustince Sola, Humberto; Herrera, Francisco; Automática y Computación; Automatika eta Konputazioa
This paper reviews the fingerprint classification literature looking at the problem from a double perspective. We first deal with feature extraction methods, including the different models considered for singular point detection and for orientation map extraction. Then, we focus on the different learning models considered to build the classifiers used to label new fingerprints. Taxonomies and classifications for the feature extraction, singular point detection, orientation extraction and learning methods are presented. A critical view of the existing literature have led us to present a discussion on the existing methods and their drawbacks such as difficulty in their reimplementation, lack of details or major differences in their evaluations procedures. On this account, an experimental analysis of the most relevant methods is carried out in the second part of this paper, and a new method based on their combination is presented.
Embargo
Non-symmetric over-time pooling using pseudo-grouping functions for convolutional neural networks
(Elsevier, 2024-07-01) Ferrero Jaurrieta, Mikel; Paiva, Rui; Cruz, Anderson; Bedregal, Benjamin; Miguel Turullols, Laura de; Takáč, Zdenko; López Molina, Carlos; Bustince Sola, Humberto; Estadística, Informática y Matemáticas; Estatistika, Informatika eta Matematika; Institute of Smart Cities - ISC
Convolutional Neural Networks (CNNs) are a family of networks that have become state-of-the-art in several fields of artificial intelligence due to their ability to extract spatial features. In the context of natural language processing, they can be used to build text classification models based on textual features between words. These networks fuse local features to generate global features in their over-time pooling layers. These layers have been traditionally built using the maximum function or other symmetric functions such as the arithmetic mean. It is important to note that the order of input local features is significant (i.e. the symmetry is not an inherent characteristic of the model). While this characteristic is appropriate for image-oriented CNNs, where symmetry might make the network robust to image rigid transformations, it seems counter-productive for text processing, where the order of the words is certainly important. Our proposal is, hence, to use non-symmetric pooling operators to replace the maximum or average functions. Specifically, we propose to perform over-time pooling using pseudo-grouping functions, a family of non-symmetric aggregation operators that generalize the maximum function. We present a construction method for pseudo-grouping functions and apply different examples of this family to over-time pooling layers in text-oriented CNNs. Our proposal is tested on seven different models and six different datasets in the context of engineering applications, e.g. text classification. The results show an overall improvement of the models when using non-symmetric pseudo-grouping functions over the traditional pooling function.
Open Access
Operador de comparación de elementos multivaluados basado en funciones de equivalencia restringida
(Universidad de Málaga, 2021) Castillo López, Aitor; López Molina, Carlos; Fernández Fernández, Francisco Javier; Sesma Sara, Mikel; Bustince Sola, Humberto; Estadística, Informática y Matemáticas; Estatistika, Informatika eta Matematika
En este trabajo proponemos un nuevo enfoque del algoritmo de clustering gravitacional basado en lo que Einstein considero su 'mayor error': la constante cosmológica. De manera similar al algoritmo de clustering gravitacional, nuestro enfoque está inspirado en principios y leyes del cosmos, y al igual que ocurre con la teoría de la relatividad de Einstein y la teoría de la gravedad de Newton, nuestro enfoque puede considerarse una generalización del agrupamiento gravitacional, donde, el algoritmo de clustering gravitacional se recupera como caso límite. Además, se desarrollan e implementan algunas mejoras que tienen como objetivo optimizar la cantidad de iteraciones finales, y de esta forma, se reduce el tiempo de ejecución tanto para el algoritmo original como para nuestra versión.
Open Access
A framework for active contour initialization with application to liver segmentation in MRI
(Springer, 2022) Mir Torres, Arnau; Antunes dos Santos, Felipe; Fernández Fernández, Francisco Javier; López Molina, Carlos; Estadística, Informática y Matemáticas; Estatistika, Informatika eta Matematika
Object segmentation is a prominent low-level task in image processing and computer vision. A technique of special relevance within segmentation algorithms is active contour modeling. An active contour is a closed contour on an image which can be evolved to progressively fit the silhouette of certain area or object. Active contours shall be initialized as a closed contour at some position of the image, further evolving to precisely fit to the silhouette of the object of interest. While the evolution of the contour has been deeply studied in literature [5, 11], the study of strategies to define the initial location of the contour is rather absent from it. Typically, such contour is created as a small closed curve around an inner position in the object. However, literature contains no general-purpose algorithms to determine those inner positions, or to quantify their fitness. In fact, such points are frequently set manually by human experts, hence turning the segmentation process into a semi-supervised one. In this work, we present a method to find inner points in relevant object using spatial-tonal fuzzy clustering. Our proposal intends to detect dominant clusters of bright pixels, which are further used to identify candidate points or regions around which active contours can be initialized.
Open Access
Hyperspectral imaging using notions from type-2 fuzzy sets
(Springer, 2019) López Maestresalas, Ainara; Miguel Turullols, Laura de; López Molina, Carlos; Arazuri Garín, Silvia; Bustince Sola, Humberto; Jarén Ceballos, Carmen; Ingeniería; Ingeniaritza; Estadística, Informática y Matemáticas; Estatistika, Informatika eta Matematika; Universidad Pública de Navarra / Nafarroako Unibertsitate Publikoa
Fuzzy set theory has developed a prolific armamentarium of mathematical tools for each of the topics that has fallen within its scope. One of such topics is data comparison, for which a range of operators has been presented in the past. These operators can be used within the fuzzy set theory, but can also be ported to other scenarios in which data are provided in various representations. In this work, we elaborate on notions for type-2 fuzzy sets, specifically for the comparison of type-2 fuzzy membership degrees, to create function comparison operators. We further apply these operators to hyperspectral imaging, in which pixelwise data are provided as functions over a certain energy spectra. The performance of the functional comparison operators is put to the test in the context of in-laboratory hyperspectral image segmentation.
Open Access
Reduction of complexity using generators of pseudo-overlap and pseudo-grouping functions
(2024) Ferrero Jaurrieta, Mikel; Paiva, Rui; Cruz, Anderson; Bedregal, Benjamin; Zhang, Xiaohong; Takáč, Zdenko; López Molina, Carlos; Bustince Sola, Humberto; Estadística, Informática y Matemáticas; Estatistika, Informatika eta Matematika
Overlap and grouping functions can be used to measure events in which we must consider either the maximum or the minimum lack of knowledge. The commutativity of overlap and grouping functions can be dropped out to introduce the notions of pseudo-overlap and pseudo-grouping functions, respectively. These functions can be applied in problems where distinct orders of their arguments yield different values, i.e., in non-symmetric contexts. Intending to reduce the complexity of pseudo-overlap and pseudo-grouping functions, we propose new construction methods for these functions from generalized concepts of additive and multiplicative generators. We investigate the isomorphism between these families of functions. Finally, we apply these functions in an illustrative problem using them in a time series prediction combined model using the IOWA operator to evidence that using these generators and functions implies better performance.
Open Access
Proyecto Agroinc: prevención del impacto ambiental de incendios provocados por cosechadoras
(Interempresas Media, 2022) Arazuri Garín, Silvia; Mangado Ederra, Jesús; López Maestresalas, Ainara; López Molina, Carlos; Angulo Muñoz, Blanca; Arnal Atarés, Pedro; Jarén Ceballos, Carmen; Ingeniería; Ingeniaritza; Estadística, Informática y Matemáticas; Estatistika, Informatika eta Matematika; Proyectos e Ingeniería Rural; Landa Ingeniaritza eta Proiektuak; Institute on Innovation and Sustainable Development in Food Chain - ISFOOD; Gobierno de Navarra / Nafarroako Gobernua
Las cosechadoras de cereales, por las condiciones ambientales en las que trabajan, alta temperatura y baja humedad, tanto ambiental como del producto que están cosechando, pueden provocar accidentalmente incendios durante la época de recolección. Los daños económicos y medioambientales que estos incendios suponen pueden ser muy importantes, ya que las condiciones de propagación del fuego son óptimas. Los principales objetivos de este proyecto han sido evaluar el impacto ambiental de los incendios producidos en Navarra en los últimos años y establecer una guía de buenas prácticas para su prevención.