Automatic vehicle-type classification plays an imperative role in the development of efficient Intelligent Transportation Systems (ITS). In this paper, a super-learner ensemble is proposed for the vehicle-type classification problem. A densely connected single-split super learner is utilized to exploit the strengths and diminish the weaknesses of the individual base learners ResNet50, Xception, and DenseNet. The super learner aims to learn fusion weights in a data-adaptive manner to obtain the optimal combination of the base learners. The proposed method is simple, robust, and enhances the discrimination capabilities among the similarly-looking classes without requiring any hand-crafted features or logical reasoning. The proposed method is evaluated using two of the most challenging publicly available traffic surveillance datasets: the MIOvision Traffic Camera Dataset (MIO-TCD) and the Beijing Institute of Technology's (BIT) vehicle classification dataset. Three variants of the super learner ensemble: RXD-CV-CW, RXD-CV-CW-NCW and Augmented-RXD, were examined on the MIO-TCD dataset with variations in applying class weights and data augmentation during training. RXD-CV-CW-NCW and Augmented-RXD share the third place among the published state-of-the-art methods reported in the MIO-TCD classification challenge. Augmented-RXD generalizes to the classes in common between the two datasets without degrading its performance on the MIO-TCD dataset. Both variants achieved an overall accuracy of 97.94%, and a Cohen Kappa score of 96.78%. In addition, the super-learner variants that we trained on the BIT-Vehicle dataset images achieved overall accuracies of up to 97.62%. INDEX TERMS Deep learning, ensemble learning, intelligent transport systems, vehicle classification. • We present a super-learner ensemble model for vehicle-type classification in surveillance frames. The super learner consists of a fully-connected layer added to the fused outputs of three base learners: ResNet50 [10], Xception [11], and DenseNet [12].
Several of the most fundamental transformations widely used by the power engineering community are strongly based on geometrical considerations. Clifford's geometric algebra (GA) is the natural language for describing concepts in Euclidean geometry. In this work, we show how Clarke, Park, and Depenbrock's FBD transformations can be derived by imposing orthogonality on the voltage and current vectors defined in a Euclidean space by using GA. This paper presents these transformations as spatial‐like rotations and projections by means of the use of special algebraic objects named GA rotors. We prove that there is no need to use complex numbers nor matrices to perform the mentioned transformations and to provide geometrical intuition for electrical quantities and their transformations. Furthermore, power properties can be described using GA terminology by means of the proposed geometric power. The manipulation of the geometric power allows a useful current decomposition for a variety of applications such active filtering, frequency estimation, or electrical machinery. We provide in this paper an alternative approach focused on power systems under the paradigm of GA.
This paper presents a new framework for a systematic and thorough generalization of the most well-known instantaneous transformations used in electrical engineering for power systems analysis and computing through geometric principles based on the language of Geometric Algebra. By introducing the concepts of Kirchhoff Vector and Kirchhoff Subspace, a new generalized transformation is presented. Thus, it is shown how the Clarke, Park or Hyper-Space vector transformations (widely used in electrical engineering) are particular cases of this unifying framework. Moreover, a generalization to an arbitrary number of phases is achieved. In order to be as close as possible to the geometrical intuition, all the underlying ideas are presented by means of spatial-like conceptualizations, substantiated by their corresponding algebraic formulation. This proposal has potential uses in a wide range of power system applications such as electrical machines, current compensation, power quality, electronic converters or transmission lines. Preliminary results show the superior efficiency of the method compared to matrix methods. Some real-world examples have been included to highlight the potential use of the method.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.