Suchergebnisse

Currently, state-of-the-art methods for 3D object recognition rely in a deep learning-pipeline. Nonetheless, these methods require a large amount of data that is not easy to obtain. In addition to that, the majority of them exploit features of the datasets, like the fact of being CAD models to create rendered representation which will not work in real life because the 3D sensors provide point clouds. We propose a novel global descriptor for point clouds which takes advantage of the fractal dimension of the objects. Our approach introduces many benefits, such as being agnostic to the density of points of the sample, number of points in the input cloud, sensor of choice, and noise up to a level, and it works on real life point cloud data provided by commercial sensors. We tested our descriptor for 3D object recognition using ModelNet, which is a well-known dataset for that task. Our approach achieves 92.84% accuracy on the ModelNet10, and 88.74% accuracy on the ModelNet40. ; This work was supported in part by the Spanish Government, with Feder funds, under Grant PID2019-104818RB-I00, and in part by the Spanish Grants for Ph.D. studies under Grant ACIF/2017/243 and Grant FPU16/00887.

Relevance feedback schemes use a given set of descriptors and at each feedback iteration they just modify the importance of each descriptor to the retrieval performance. This implies that the same type of descriptors is used for all feedback iterations. Therefore, such approaches cannot improve the retrieval performance beyond the efficiency of the descriptors used to represent the multimedia content. In this paper, an alternative concept is to adopt a progressive descriptor organization and to allow dynamic modification of the type of descriptors (expansion or shrink) at each retrieval iteration by exploiting the current user's information needs. In other words, descriptors that best fit the current user's preferences are expanded into more details, in constant to the descriptors that are far way from the user's needs which are shrinking. In this case, instead of using a large set of descriptors, most of them inappropriate with respect to the current user's information needs, a small number of dominant descriptors is selected for the retrieval process. Thus, the proposed approach significantly increases the precision performance by discarding descriptors that yield noise in the retrieval. © 2007 IEEE.

Trabajo presentado al 20th IFAC (International Federation of Automatic Control) World Congress, celebrado en Toulouse (Francia) del 9 al 14 de julio de 2017. ; This paper considers actuator-fault estimation for discrete-time descriptor systems with unknown but bounded system disturbance and measurement noise. A zonotopic fault estimation filter is designed based on the analysis of fault detectability indexes. To ensure estimation accuracy, the filter gain in the zonotopic fault estimation filter is optimized through the zonotope minimization. The designed zonotopic filter not only can estimate fault magnitudes, but it also provides fault estimation results in an interval, i.e. the upper and lower bounds of fault magnitudes. Moreover, the proposed fault estimation filter has a non-singular structure and hence is easy to implement. Finally, simulation results are provided to illustrate the effectiveness of the proposed method. ; This work has been partially funded by the Spanish Government and FEDER through the projects CICYT ECOCIS (ref. DPI2013-48243-C2-1-R), CICYT DEOCS (ref. DPI2016-76493-C3-3-R), CICYT HARCRICS (ref. DPI2014-58104-R) and by National Natural Science Foundation of China (Grant No. 61273162, 61403104). ; Peer Reviewed

Total pressure distortion is one of the three basic flow distortions (total pressure, total temperature and swirl distortion) that might appear at the inlet of a gas turbine engine (GTE) during operation. Different numerical parameters are used for assessing the total pressure distortion intensity and extent. These summary descriptors are based on the distribution of total pressure in the aerodynamic interface plane. There are two descriptors largely spread around the world, however, three or four others are still in use and can be found in current references. The staff at the University of Defence decided to compare the most common descriptors using basic flow distortion patterns in order to select the most appropriate descriptor for future department research. The most common descriptors were identified based on their prevalence in widely accessible publications. The construction and use of these descriptors are reviewed in the paper. Subsequently, they are applied to radial, angular, and combined distortion patterns of different intensities and with varied mass flow rates. The tests were performed on a specially designed test bench using an electrically driven standalone industrial centrifugal compressor, sucking air through the inlet of a TJ100 small turbojet engine. Distortion screens were placed into the inlet channel to create the desired total pressure distortions. Of the three basic distortions, only the total pressure distortion descriptors were evaluated. However, both total and static pressures were collected using a multi probe rotational measurement system.

This article discusses assessing the instability of a continuous linear homogeneous timeinvariant descriptor system. Some necessary conditions and sufficient conditions are derived to establish the stability of a matrix pair by the fundamentals of qualitative ecological principles. The proposed conditions are derived using only the qualitative (sign) information of the matrix pair elements. Based on these conditions, the instability of a matrix pair can easily be determined, without any magnitude information of the matrix pair elements and without numerical eigenvalues calculations. With the proposed theory, Magnitude Dependent Stable, Magnitude Dependent Unstable, and Qualitative Sign Stable matrix pairs can be distinguished. The consequences of the proposed conditions and some illustrative examples are discussed.

El reconocimiento de gestos se ha presentado como una alternativa para la implementación de sistemas de interacción eficaces. Particularmente las aplicaciones basadas en visión artificial poseen ventajas en portabilidad frente a otras alternativas. Sin embargo, los algoritmos suelen requerir entrenamiento computacional intensivo, siendo difíciles de implementar en dispositivos móviles. En este artículo se realiza un estudio preliminar para la detección de poses de la mano usando un algoritmo basado en Patrones Binarios locales, más conocido por su sigla en inglés LBP (Local Binary Patterns). Para lo anterior, se presenta un modelo heurístico de división de la mano en regiones ponderadas diferencialmente, que permite la clasificación directa de los gestos usando una medida de similitud. Los pesos y la distribución de regiones se evaluaron de acuerdo a su precisión en la clasificación de cada pose de la mano. Estas pruebas se realizaron en un conjunto de imágenes, capturadas en condiciones controladas, correspondientes a cinco poses diferentes. El algoritmo propuesto con el esquema de regiones ponderadas muestra una buena capacidad de discriminación y presenta una alternativa válida para futuras aplicaciones.

This article deals with the 2D image-based recognition of industrial parts. Methods based on histograms are well known and widely used, but it is hard to find the best combination of histograms, most distinctive for instance, for each situation and without a high user expertise. We proposed a descriptor subset selection technique that automatically selects the most appropriate descriptor combination, and that outperforms approach involving single descriptors. We have considered both backward and forward mechanisms. Furthermore, to recognize the industrial parts a supervised classification is used with the global descriptors as predictors. Several class approaches are compared. Given our application, the best results are obtained with the Support Vector Machine with a combination of descriptors increasing the F1 by 0.031 with respect to the best descriptor alone. ; This paper has been supported by the project SHERLOCK under the European Union's Horizon 2020 Research & Innovation programme, grant agreement No. 820689.

Component: 1.1 Initial Assessment Sub-component: 1.1.3 Marine Environment Annex The purpose of this document is to identify the specific ecological characteristics of the marine management area. It aims at identifying where the particularly sensitive or ecologically important areas are. Identifying where ecologically or biologically significant areas are located is a key step for marine spatial planning. This assessment is centered on the Marine Strategy Framework Directive (European Parliament et Council of European Union s. d.) 'Good Environmental Status' Descriptor 1 (Biodiversity) and Descriptor 4 (Ecosystems, including food webs). ; This report was produced as part of SIMNORAT Project (Grant Agreement N0. EASME/EMFF/2015/1.2.1.3/03/SI2.742089). Competition for maritime space – for renewable energy equipment, aquaculture and other uses – has highlighted the need to manage our waters more coherently. Maritime spatial planning (MSP) works across borders and sectors to ensure human activities at sea take place in an efficient, safe and sustainable way. That is why the European Parliament and the Council have adopted a legislation to create a common framework for maritime spatial planning in Europe. The Directive 2014/89/EU of the European Parliament and of the Council of 23 July 2014 (said Maritime Spatial Planning Directive) establishes a framework in order to reduce conflicts between sectors and create synergies between different activities, to encourage investment – by creating predictability, transparency and clearer rules, to increase cross-border cooperation – between EU countries to develop energy grids, shipping lanes, pipelines, submarine cables and other activities, but also to develop coherent networks of protected areas, and to protect the environment – through early identification of impact and opportunities for multiple use of space. The SIMNORAT project (Supporting Implementation of Maritime Spatial Planning in the Northern European Atlantic) is an EU/DG Mare co-funded cross-border project. It ...