In this paper, we propose a novel method that performs dynamic action classification by exploiting the effectiveness of the Extreme Learning Machine (ELM) algorithm for single hidden layer feedforward neural networks training. It involves data grouping and ELM based data projection in multiple levels. Given a test action instance, a neural network is trained by using labeled action instances forming the groups that reside to the test sample’s neighborhood. The action instances involved in this procedure are, subsequently, mapped to a new feature space, determined by the trained network outputs. This procedure is performed multiple times, which are determined by the test action instance at hand, until only a single class is retained. Experimental results denote the effectiveness of the dynamic classification approach, compared to the static one, as well as the effectiveness of the ELM in the proposed dynamic classification setting. 相似文献
In this paper, a human action recognition method is presented in which pose representation is based on the contour points of the human silhouette and actions are learned by making use of sequences of multi-view key poses. Our contribution is twofold. Firstly, our approach achieves state-of-the-art success rates without compromising the speed of the recognition process and therefore showing suitability for online recognition and real-time scenarios. Secondly, dissimilarities among different actors performing the same action are handled by taking into account variations in shape (shifting the test data to the known domain of key poses) and speed (considering inconsistent time scales in the classification). Experimental results on the publicly available Weizmann, MuHAVi and IXMAS datasets return high and stable success rates, achieving, to the best of our knowledge, the best rate so far on the MuHAVi Novel Actor test. 相似文献
The performance of recent dereverberation methods for reverberant speech preprocessing prior to Automatic Speech Recognition (ASR) is compared for an extensive range of room and source-receiver configurations. It is shown that room acoustic parameters such as the clarity (C50) and the definition (D50) correlate well with the ASR results. When available, such room acoustic parameters can provide insight into reverberant speech ASR performance and potential improvement via dereverberation preprocessing. It is also shown that the application of a recent dereverberation method based on perceptual modelling can be used in the above context and achieve significant Phone Recognition (PR) improvement, especially under highly reverberant conditions. 相似文献
Clustering of nodes in optical networks has been proven to be an efficient way to serve end-to-end connectivity. However, clustering requires specific topological characteristics, or alternatively the introduction of significant alterations of an existing topology to achieve the expected performance improvements. The comparison of future dynamic optical networking technologies should therefore include in the set of initial assumptions, apart from the statistical properties of the traffic load, the network topology to draw conclusions regarding the efficiency as well as feasibility and scalability of the proposed solutions. In this article, we show how node clustering under the CANON architecture can be applied in real-life core networks and provide superior performance compared to conventional burst switching techniques in terms of blocking, resource utilization and power consumption. 相似文献
In this paper we present a new location management protocol for wireless ATM networks, called LMCP (Location Management and Control Protocol). This protocol is based on the PNNI (Private Network-to-Network Interface) routing functionality to advertise the movement of mobile terminals within predefined areas. Moreover, LMCP uses specialized entities to store and retrieve the current location area of the mobile terminals. These entities are located in mobility enhanced switches that control the execution of mobility procedures (e.g., handovers).The main benefit from the application of LMCP is the establishment of connections that do not contain any misrouted segments. Furthermore, it requires minor modifications to the PNNI and enables its inter-working with other location management mechanisms. The protocol is compared with other similar mechanisms, and its efficiency is demonstrated by the results of an analytical model. 相似文献
A current premium vehicle is implemented with a variety of information, entertainment, and communication functions, which are generally referred as an infotainment system. During vehicle development, testing of the infotainment system at an overall level is conventionally carried out manually by an expert who can observe at a customer level. This approach has significant limitations with regard to test coverage and effectiveness due to the complexity of the system functions and human’s capability. Hence, it is highly demanded by car manufacturers for an automated infotainment testing system, which replicates a human expert encompassing relevant sensory modalities relating to control (i.e., touch) and observation (i.e., sight and sound) of the system under test. This paper describes the design, development, and evaluation of such a system that consists of simulation of vehicle network, vision-based inspection, automated navigation of features, random cranking waveform generation, sound detection, and test automation. The system developed is able to: stimulate a vehicle system across a wide variety of initialisation conditions, exercise each function, check for system responses, and record failure situations for post-testing analysis. 相似文献
This paper provides a concise overview of the influence of human activity within the UK society on resource consumption and the subsequent effects on the environment. The concept of the Multi Scale Integrated Analysis of Societal Metabolism (MSIASM) is applied in order to elucidate the evolution of the UK economy for the period between 1981 and 2004. Our findings highlight the transition to a service-based economy and the disproportionate increase of energy demand when compared to the overall population increase. Emergy synthesis is applied in order to understand the production and consumption patterns and the environmental support required to sustain human activity within the UK for the year 2004. Generally speaking the UK society greatly benefits from its significant natural resources with 44.3% of the total emergy used coming from home sources and 29.1% from locally renewable sources. Interestingly enough, despite its significant natural resources, the UK economy, seems to be a net emergy importer by 638.5 × 1021 seJ. Furthermore, the current economic activity is believed to have a significant impact in the environment despite the relatively low environmental load ratio of 2.44. 相似文献
Recycling of PET was examined using hydrolytic depolymerization in an alkaline solution under microwave irradiation. The reaction was carried out in a sealed microwave reactor in which the pressure and temperature were controlled and recorded. The main products were the monomers TPA and EG. The effect of reaction temperature, time, amount of PET and alkaline concentration on the degree of PET depolymerization and TPA recovery was investigated. Microwave irradiation was found to reduce the time needed to achieve a specific degradation of PET significantly, with almost complete depolymerization occurring in 30 min at 180 °C and only 46 W of microwave power. Using a phase transfer catalyst (TOMAB) resulted in the same amount of unreacted PET but at significantly lower depolymerization temperatures.
It has very often been stated that the difficulty and complexity of achieving green energy targets in the European Union (EU) will require strengthened measures to promote implementation of New Energy Technologies, Energy End-use Efficiency, as well as measures to support the related energy Research and Technology Development (RTD). Often forgotten is the fact, that most of all, a European-wide co-ordinated forum is needed to continuously develop and sophisticate the monitoring and methodology results, bringing together specialised statisticians, energy researchers and experts on energy socio-economics. The aim of this paper is to present the Scientific Reference System (SRS) Scorecard; a data validation platform for the sophisticated monitoring and communication of the energy technology sector. In this respect, the concept of the SRS scorecard system will be laid out, the parameters and the scoring criteria will be explained as well as the assessment system so as to provide the interested reader with the basis needed to understand the technology evaluation examples provided, as well as its critical analysis. 相似文献