Image shadow segmentation has become a major issue in satellite remote sensing because of the recent commercial availability high-resolution images. Detecting shadows is important for successfully carrying out applications such as change detection, land monitoring, object recognition, scene reconstruction, colour correction, etc. This paper presents simple and effective procedure to segment regions on The method applies region growing process specific band (namely, c 3 component 1 2 space)....

This paper presents a new method for recognizing places in indoor environments based on the extraction of planar regions from range data provided by hand-held RGB-D sensor. We propose to build plane-based map (PbMap) consisting set 3D patches described simple geometric features (normal vector, centroid, area, etc.). world representation is organized as graph where nodes represent and edges connect planes that are close by. structure permits efficiently select subgraphs representing local...

The integration of several range cameras in a mobile platform is useful for applications robotics and autonomous vehicles that require large field view. This situation increasingly interesting with the advent low cost like those developed by Primesense. Calibrating such combination sensors any geometric configuration problem has been recently solved through visual odometry (VO) SLAM. However, this kind solution laborious to apply, requiring robust SLAM or VO controlled environments. In paper...

Many applications in the fields of mobile robotics and autonomous vehicles employ two or more 2D laser rangefinders (LRFs) for different purposes: navigation, obstacle detection, 3D mapping simultaneous localization mapping. The extrinsic calibration between such sensors (i.e. finding their relative poses) is required to exploit effectively all sensor measurements perform data fusion. In literature, most works employing several LRFs obtain from manual ad-hoc solutions. this paper we present...

In remote sensing, because of the wide diversity image characteristics (size, spatial and radiometric resolution, terrain relief, observation poses, etc.), registration methods that work well on certain satellite images may not produce acceptable results for others, requiring more powerful techniques. A variety techniques account with non‐rigid geometric deformations have been proposed, including piecewise (linear or cubic) functions, weighted mean radial basis B‐spline etc. This paper...

The integration of several 2D laser rangefinders in a vehicle is common resource employed for 3D mapping, obstacle detection and navigation. extrinsic calibration between such sensors (i.e. finding their relative poses) required to exploit effectively the sensor measurements perform data fusion. approaches found literature obtain either rely on approximated parameters from rig construction or propose ad-hoc solutions specific LRF rigs. In this paper we present novel solution set at least...

For mobile robots to operate in an autonomous and safe manner they must be able adequately perceive their environment despite challenging or unpredictable conditions sensory apparatus. Usually, this is addressed through ad-hoc, not easily generalizable Fault Detection Diagnosis (FDD) approaches. In work, we leverage Bayesian Networks (BNs) propose a novel probabilistic inference architecture that provides generality, rigorous inferences real-time performance for the detection, diagnosis...

Piecewise linear transformation is a powerful technique for coping with the registration of images affected by local geometric distortions, as it usually case high-resolution satellite images. A key point when applying this to divide register according suitable common triangular mesh. This comprises two different aspects: where place mesh vertices (i.e., geometrical realization) and set an appropriate topology upon these topological realization). paper focuses on latter presents novel method...

<p class="icsmabstract">En este trabajo se presenta una iniciativa encaminada a introducir y fomentar el uso de los videotutoriales como herramienta apoyo para aprendizaje conceptos básicos del control automático en titulaciones universitarias técnicas. Los contenidos desarrollados, disponibles canal YouTube denominado “Automática”, han agrupado acuerdo principales bloques conceptuales la materia sido puestos disposición alumnado espacio campus virtual Universidad Málaga (UMA) asociado...

This paper describes SANCHO, a mobile robot intended to perform within crowded areas as servant, for instance fair or congress host. has been constructed upon commercial platform on which number of sensors and devices have integrated. A software control architecture implemented adapted this particular robot, enabling it in human scenarios. Among the different subsystems developed we highlight two most relevant ones: navigation component permits navigate safe robust manner dynamic...

Detecting abnd localizing objects from space, like roads, rivers, lakes, etc., is a challenging task with multiple applications in remote sensing. In this paper we address the detection of swimming pools colour high-resolution images urban areas acquired by Quickbird satellite. The main motivation work to survey and localize filled during drought periods, fact that should be then punished local authorities. proposed algorithm applies analysis for water approximate segmentation, as an...

Abstract Setting‐out is a crucial and frequently repeated process in construction civil engineering. It carried out by qualified operators who, making use of surveying stations, identify reference points to guide workers their tasks. In this paper, we focus on the particular case setting‐out road/train tunnels propose system that automatically performs operation tunnel section be perforated. The presented system, called Tunnel Continuous Setout (TCS), integrates scanning device surveys...

In the last very few years, monocular SLAM approaches based on bundle adjustment are achieving amazing results in terms of accuracy, computational efficiency, and density map. When such solutions applied large scenarios it is crucial for system scalability to maintain a map representation that permits efficient optimization augmentation. order cope with maps, we present an on-the-fly partitioning technique which allows abstraction from metric operate more efficiently. The result...

The representation of the world upon planar patches has proven to be simple, robust and useful for a variety robotic tasks, including SLAM, autonomous navigation, or scene recognition. In this work we investigate how incorporate color information into such improve matching while maintaining model compactness, which is essential real-time applications. We propose descriptor based on dominant patch, defined as center biggest cluster in patch histogram. paper, different spaces methods...

El número de personas con problemas movilidad está creciendo en los últimos años países desarrollados, como consecuencia una mayor esperanza vida y del gran accidentes tráfico. La posibilidad transferir adaptar resultados campo la robótica a este colectivo resultaría enorme interés toda vez que repercuta mejor calidad integración social. En artículo presentamos silla ruedas robotizada SENA, es el resultado un proyecto investigación desarrollado Dpto. Ingeniería Sistemas Automática...

The availability of increasing satellite image resolution may help city administrations in the daily management urban development and planning. In this paper we describe a software system, called DECASAT, that automatically detects changes between two images taking from same scene. This system has been tested successfully with panchromatic provided by IRS 1D. We also show how resulting change is integrated into GIS Malaga (Spain).

En este trabajo se propone una arquitectura de reflejos ante estímulos sensoriales para que el aprendizaje por refuerzo en vivo robots móviles mejore su adaptación a cambios la tarea, aumente autonomía regresar estados seguros tras errores y reduzca, general, supervisión parte del humano. El ha enfocado navegación un robot móvil con evitación obstáculos hemos utilizado versión modificada los algoritmos Q-learning True On-Line SARSA (lambda). Se estudiado adicionalmente traslada lo aprendido...

Este trabajo presenta un sistema basado en visión que utiliza redes neuronales para la estimación de poses humanas 3D. La solución desarrollada identifica el sujeto analizado 18 "puntos clave" o keypoints mediante cuatro cámaras RGB calibradas. utilización múltiples permite superar problemas inherentes al uso una sola cámara RGBD/estéreo como pérdida por existencia oclusiones mayor incertidumbre profundidad, proporcionando base robusta futuras investigaciones y aplicaciones campos...