9300 research outputs found
Sort by
Centroid based person detection using pixelwise prediction of the position
Get PDFImplementations of person detection in tracking and counting systems tend towards processing of orthogonally captured images on edge computing devices. The ellipse-like shape of heads in orthogonally captured images inspired us to predict head centroids to determine positions of persons in images. We predict the centroids using a fully convolutional network (FCN). We combine the FCN with simple image processing operations to ensure fast inference of the detector. We experiment with the size of the FCN output to further decrease the inference time. We compare the proposed centroid-based detector with bounding box-based detectors on head detection task in terms of the inference time and the detection performance. We propose a performance measure which allows quantitative comparison of the two detection approaches. For the training and evaluation of the detectors, we form original datasets of 8000 annotated images, which are characterized by high variability in terms of lighting conditions, background, image quality, and elevation profile of scenes. We propose an approach which allows simultaneous annotation of the images for both bounding box-based and centroid-based detection. The centroid-based detector shows the best detection performance while keeping edge computing standards.The work was supported from ERDF/ESF “Cooperation in Applied Research between the University of Pardubice and companies, in the Field of Positioning, Detection and Simulation Technology for Transport Systems (PosiTrans)” (No. CZ.02.1.01/0.0/0.0/17_049/0008394)
Extension of locally adapted models of photosynthetically active radiation for all sky conditions
No full textComunicación presentada en: XII Congreso Nacional y III Internacional de Ingeniería Termodinámica (12 CNIT), June 19- July 1, Madrid (Spain)Photosynthetically Active Radiation (PAR, 400-700 nm) is the energy source to trigger photosynthesis. This process makes food and biomass production and forest productivity possible, so it becomes essential for determining the impact of deforestation and climate change on agriculture. Due to the scarcity of PAR data from direct measurements at ground meteorological stations, empirical models based on linear regressions have been developed for estimate PAR data, using other meteorological and climatic variables. In recent years, machine learning algorithms have been discovered as a useful tool for modelling meteorological and climatic data. Thus, Artificial Neural Networks (ANN) have been used for modelling PAR, with different meteorological variables as input. Both procedures, multilinear regressions and ANN’s, have been used in this work for modelling PAR in Burgos (Spain) under all sky conditions attending to the sky clearness classification and in an hourly basis. The performance of the resulting models has been tested for PAR estimates at other locations. To this end, he experimental data obtained from the Surface Radiation Budget Network (SURFRAD) in the USA was used. This proves the good fit of the models developed in Burgos to the SURFRAD weather stations.Financial support was provided by the Spanish MCIN (Ref. RTI2018-098900-B-I00). Junta de Castilla y León provided financial support for Diego Granados López and Elena Garrachón Gómez (ORDEN EDU/556/2019 and Programa Operativo de Empleo Juvenil, Fondo Social Europeo, respectively)
Dataset of the paper “A stepped tabu search method for the clique partitioning problem”. Applied Intelligence, 53, 16275-16292
No full textGiven an undirected graph, a clique is a subset of vertices in which the induced subgraph is complete; that is, all pairs of vertices of this subset are adjacent. Clique problems in graphs are very important due to their numerous applications. One of these problems is the clique partitioning problem (CPP), which consists of dividing the set of vertices of a graph into the smallest number of cliques possible. The CPP is an NP-hard problem with many application fields (timetabling, manufacturing, scheduling, telecommunications, etc.). Despite its great applicability, few recent studies have focused on proposing specific resolution methods for the CPP. This article presents a resolution method that combines multistart strategies with tabu search. The most novel characteristic of our method is that it allows unfeasible solutions to be visited, which facilitates exploration of the solution space. The computational tests show that our method performs better than previous methods proposed for this problem. In fact, our method strictly improves the results of these methods in most of the instances considered while requiring less computation time.This work was partially supported by FEDER funds and the Spanish State Research Agency (Projects PID2019-104263RB-C44 and PDC2021–121021-C22); the Regional Government of “Castilla y León”, Spain (Project BU071G19); the Regional Government of “Castilla y León”; and FEDER funds (Project BU056P20)
Kahoot!: Effective, simple and easy gamification in Higher Education: practical application in an Engineering Degree
No full textEl Espacio Europeo de Educación Superior (EEES) establece que el aprendizaje debe estar centrado en el estudiante. Partiendo de este enunciado, el artículo explica cómo la gamificación puede ser una técnica pedagógica que propicie el ansiado cambio en la docencia universitaria hacia donde marca el EEES. Este trabajo presenta el diseño de una metodología de gamificación sencilla y fácil de implementar en un Grado de Ingeniería utilizando Kahoot! como herramienta y su evolución a lo largo de tres cursos académicos. Los resultados indican que los alumnos ven más atractiva la asignatura, aumenta el seguimiento y mejoran sus resultados académicos.The European Higher Education Area (EHEA) establishes that learning must be
student-centered. Based on this statement, the article explains how gamification
can be a pedagogical technique that can provide the desired change in university
teaching towards the EHEA patterns. This paper presents the design of a simple
and easy to implement gamification methodology in an Engineering Degree using
Kahoot! as a tool and its evolution over three academic years. The results indicate
that the students find the subject more attractive, its follow-up increases and their
academic results improve