Search

Accurate and robust localization is one of the most basic functional modules of mobile robots. A recent promising approach is to employ a camera that captures ground textures to achieve high-precision localization. However, low robustness and real-time performance are disadvantages of the system due to the time-consuming front-end feature extraction and search methods. To solve these problems, a comprehensive system that has high robustness and real-time performance is proposed, including offline mapping, global localization, local localization and map update. This paper summarizes a method for selecting the optimal front-end for different textures. A fast global localization method based on bag of words is proposed, which can quickly and effectively determine the local positioning area. Experiments are carried out on public datasets and actual indoor scenes to verify the practicability of the system. [ABSTRACT FROM AUTHOR]

During recent years, deep convolutional neural networks (CNNs) have become increasingly popular in sensor-based human activity recognition (HAR), due to powerful feature extraction capacity. However, the required high computational cost has hindered their practical deployment in real-world applications. To address this issue, this paper presents a novel approach named ABCSearch to prune channels through an artificial bee colony (ABC) algorithm, which is a meta-heuristic optimization algorithm inspired by the foraging behavior of honeybees. Unlike traditional pruning strategies that prioritize channel importance, the proposed method seeks to identify the optimal number of channels at each layer, i.e., the optimal pruning structure. To lessen human interference, we transform the search problem of the optimal pruning structure into a simple optimization problem and solve it automatically through the ABC algorithm, which can significantly decrease both model size and computational cost without compromising performance. Specifically, we validate the effectiveness of ABCSearch for a standard five-layer CNN backbone architecture, which achieves 97.05%, 98.45%, 91.71%, 78.89% classification accuracies on four HAR benchmarks, i.e., UCI-HAR, WISDM, PAMAP2, UniMiB-SHAR. Our approach can yield 87.44%, 89.41%, 89.06%, 89.59% FLOPs reductions, and 89.54%, 88.70%, 88.38%, 90.45% less memory consumptions, as well as 4.4 × , 5.0 × , 5.8 × , 6.1 × practical speedups on real hardware devices, while maintaining comparable accuracy. [ABSTRACT FROM AUTHOR]

It is one of the hot topics in the field of computer vision to estimate the density of vehicles on the road by using UAV equipped camera. Because the vehicle scale in video images is variable, the scene is complex and the change of vehicle density has strong randomness, the estimation results of vehicle density obtained by the existing methods have large errors. To solve these problems, this paper proposes a vehicle density estimation network based on multi-scale sensing of vehicle density (DSNet). The network adaptively senses the vehicle density in video images through an innovatively designed density sensing module and generates feature maps with different densities. Combined with the context information extracted from the network, the appropriate weight for each density-specific feature map is generated. By fusing the feature maps with different weights to enhance the feature of vehicle. Finally, the enhanced feature map is further refined to obtain a high-resolution density map. In addition, to solve the problem of the global counting loss (L C) that the global estimation results are good but the local estimation errors exist, we design and add the new region-aware loss (L G) to enhance the local feature extraction ability of the network and improve the robustness of the network. Experimental results show that the GAME of DSNet on the VisDrone 2019 dataset and DETRAC dataset is reduced to 5.45 and 6.07, respectively. Compared with the current state-of-the-art vehicle density estimation methods, DSNet achieves higher accuracy in the vehicle density estimation tasks of different scenes. [ABSTRACT FROM AUTHOR]

In the not-so-far future, autonomous vehicles will be ubiquitous and, consequently, need to be coordinated to avoid traffic jams and car accidents. A failure in one or more autonomous vehicles may break this coordination, resulting in reduced efficiency (due to traffic load) or even bodily harm (due to accidents). The challenge we address in this paper is to identify the root cause of such failures. Identifying the faulty vehicles in such cases is crucial in order to know which vehicles to repair to avoid future failures as well as for determining accountability (e.g., for legal purposes). More generally, this paper discusses multi-agent systems (MAS) in which the agents use a shared pool of resources and they coordinate to avoid resource contention by agreeing on a temporal resource allocation . The problem we address, called the Temporal Multi-Agent Resource Allocation (TMARA) diagnosis problem ( TMARA-Diag ), is to find the root cause of failures in such MAS that are caused by malfunctioning agents that use resources not allocated to them. As in the autonomous vehicles example, such failures may cause the MAS to perform suboptimally or even fail, potentially causing a chain reaction of failures, and we aim to identify the root cause of such failures, i.e., which agents did not follow the planned resource allocation. We show how to formalize TMARA-Diag as a model-based diagnosis problem and how to compile it to a set of logical constraints that can be compiled to Boolean satisfiability (SAT) and solved efficiently with modern SAT solvers. Importantly, the proposed solution does not require the agents to share their actual plans, only the agreed upon temporal resource allocation and the resources used at the time of failure. Such solutions are key in the development and success of intelligent, large, and security-aware MAS. [ABSTRACT FROM AUTHOR]

Highlights • Combining matrix descriptors with vector features improves classifier performance. • 1D feature vectors treated as trees can be used to build matrix representations. • Features extracted from matrix representation of input vectors are generalizable to many problems. Abstract Pattern representation affects classification performance. Although discovering "universal" features that work for many classification problems is ideal, most representations are problem specific. In this paper, we improve the classification performance of a classifier system by transforming a one-dimensional input descriptor into a two-dimensional space so that effective texture extractors can be extracted to capture hidden data information. We develop two new methods for matrix representation where features are extracted that are more generalizable. The first method for generating a two-dimensional representation of patterns is based on trees, the objective being to reshape the feature vector into a matrix, and the second method performs mathematical operations to build a matrix representation. The proposed framework is then evaluated for its specific power on three medical problems. To evaluate generalizability, we compare the proposed approaches with several other baseline methods across some well-known benchmark datasets that reflect a diversity of classification problems. Not only does our approach show high robustness, but it also exhibits low sensitivity to parameters. When different approaches for transforming a vector into a matrix are combined with several texture descriptors, the resulting system often works well without requiring any ad-hoc optimization. The performance of the tested systems is compared by Wilcoxon signed rank test and Friedman's test. [ABSTRACT FROM AUTHOR]

Highlights • Grasshopper optimization algorithm is used to propose a feature selection method. • Computation of feature goodness factor helps to find global solution. • The accuracy of 100% is achieved by selecting 4 features out of 2308 in a dataset. • The method achieves the highest classification accuracy in 7 out of 10 datasets. Abstract Feature selection is the problem of finding the minimum number of features among a redundant feature space which leads to the maximum classification performance. In this paper, we have proposed a novel feature selection method based on mathematical model of interaction between grasshoppers in finding food sources. Some modifications were applied to the grasshopper optimization algorithm (GOA) to make it suitable for a feature selection problem. The method, abbreviated as GOFS is supplemented by statistical measures during iterations to replace the duplicate features with the most promising features. Several publicly available datasets with various dimensionalities, number of instances, and target classes were considered to evaluate the performance of the GOFS algorithm. The results of implementing twelve well-known and recent feature selection methods were presented and compared with GOFS algorithm. Comparative experiments indicate the significance of the proposed method in comparison with other feature selection methods. [ABSTRACT FROM AUTHOR]

Highlights • A new congestion function is defined as a surplus flow on hub per total flow ratio. • Particle Swarm Optimization helps dealing with the nonlinearity of the model. • The model selects more hub and tolerates higher cost to obtain homogeneous network. • Suggested solution procedure is very efficient and yield promising outcomes. • Considering time, capacity or cost factors of congestion isa visionary approach. Abstract In this paper we address the hub location problem in which capacity restrictions are introduced into the objective function as a penalty cost to represent their congestion effects on respective hubs. The goal of hub location problems is to locate an optimal set of hubs while minimizing sum of the transportation and opening costs under various constraints such as number of hubs, delay time and capacities of hubs. Moreover, we propose a more realistic variant of the problem where the capacities of hubs are not only a limitation in selecting the hubs. In practice, capacities of hubs are predicted by a strategic plan, but these predictions generally fail to meet the expectations due to changes in the flow in future and competitive nature of air transportation. A particle swarm optimization (PSO) is proposed to handle the complex nature of this problem and shorten the computing times. The performance of the proposed PSO is analyzed comparatively on the Australia Post (AP) data sets. The numerical results show that the proposed method can find the optimal solutions in less computation time in comparison with the earlier multiple allocation hub location models. [ABSTRACT FROM AUTHOR]

Highlights • We propose a new joint neural model for entity recognition and relation extraction. • Our model does not rely on external NLP tools nor hand-crafted features. • Entities and relations within the same sentence are extracted simultaneously. • We experimentally evaluate extensively on four datasets (ACE04, CoNLL04, ADE, DREC). • Our model achieves state-of-the-art performance in different contexts and languages. Abstract State-of-the-art models for joint entity recognition and relation extraction strongly rely on external natural language processing (NLP) tools such as POS (part-of-speech) taggers and dependency parsers. Thus, the performance of such joint models depends on the quality of the features obtained from these NLP tools. However, these features are not always accurate for various languages and contexts. In this paper, we propose a joint neural model which performs entity recognition and relation extraction simultaneously, without the need of any manually extracted features or the use of any external tool. Specifically, we model the entity recognition task using a CRF (Conditional Random Fields) layer and the relation extraction task as a multi-head selection problem (i.e., potentially identify multiple relations for each entity). We present an extensive experimental setup, to demonstrate the effectiveness of our method using datasets from various contexts (i.e., news, biomedical, real estate) and languages (i.e., English, Dutch). Our model outperforms the previous neural models that use automatically extracted features, while it performs within a reasonable margin of feature-based neural models, or even beats them. [ABSTRACT FROM AUTHOR]

This paper describes a strategy to identify the authorship of online handwritten documents. We regard our research framework to that of a retrieval problem and adapt the so called codebook based Vector of Local Aggregate descriptor (VLAD) that has been promising for the object retrieval application in image processing. The codebook comprises a set of code vectors with associated Voronoi cells computed from a clustering algorithm on a set of feature vectors along the online trace. However, we show that the VLAD formulation at times, cannot effectively discriminate between writers, when their respective feature vectors are not linearly separable in the Voronoi cell of the code vectors. To overcome this problem, we propose a novel descriptor that improves upon the VLAD formulation. Secondly, we explore a normalization for the feature vectors prior to the generation of the VLAD. Our method is different to the min–max and z -score in that it takes care in ensuring that the codevectors are not influenced by the presence of outliers in the data. The performance of our proposed descriptor with the new feature normalization are evaluated on two publicly available Online Handwriting Databases – the IAM and IBM-UB1. The results show a marked improvement over the VLAD. [ABSTRACT FROM AUTHOR]

Abstract: In this paper, an adequate set of input features is selected for functional expansion genetically for the purpose of solving the problem of classification in data mining using functional link neural network. The proposed method named as HFLNN aims to choose an optimal subset of input features by eliminating features with little or no predictive information and designs a more compact classifier. With an adequate set of basis functions, HFLNN overcomes the non-linearity of problems, which is a common phenomenon in single layer neural networks. The properties like simplicity of the architecture (i.e., no hidden layer) and the low computational complexity of the network (i.e., less number of weights to be learned) encourage us to use it in classification task of data mining. We present a mathematical analysis of the stability and convergence of the proposed method. Further the issue of statistical tests for comparison of algorithms on multiple datasets, which is even more essential in data mining studies, has been all but ignored. In this paper, we recommend a set of simple, yet safe, robust and non-parametric tests for statistical comparisons of the HFLNN with functional link neural network (FLNN) and radial basis function network (RBFN) classifiers over multiple datasets by an extensive set of simulation studies. [Copyright &y& Elsevier]

This paper proposes hybrid feature selection approaches based on the Genetic Algorithm (GA). This approach uses a hybrid search technique that combines the advantages of filter feature selection methods with an enhanced GA (EGA) in a wrapper approach to handle the high dimensionality of the feature space and improve categorization performance simultaneously. First, we propose EGA by improving the crossover and mutation operators. The crossover operation is performed based on chromosome (feature subset) partitioning with term and document frequencies of chromosome entries (features), while the mutation is performed based on the classifier performance of the original parents and feature importance. Thus, the crossover and mutation operations are performed based on useful information instead of using probability and random selection. Second, we incorporate six well-known filter feature selection methods with the EGA to create hybrid feature selection approaches. In the hybrid approach, the EGA is applied to several feature subsets of different sizes, which are ranked in decreasing order based on their importance, and dimension reduction is carried out. The EGA operations are applied to the most important features that had the higher ranks. The effectiveness of the proposed approach is evaluated by using naïve Bayes and associative classification on three different collections of Arabic text datasets. The experimental results show the superiority of EGA over GA, comparisons of GA with EGA showed that the latter achieved better results in terms of dimensionality reduction, time and categorization performance. Furthermore, six proposed hybrid FS approaches consisting of a filter method and the EGA are applied to various feature subsets. The results showed that these hybrid approaches are more effective than single filter methods for dimensionality reduction because they were able to produce a higher reduction rate without loss of categorization precision in most situations. [ABSTRACT FROM AUTHOR]

Novelty detection is a problem with a large number of relevant applications. For some applications, the main interest is the prevention or detection of undesired states. In some cases, these undesired states are not known in advance; in others, such as machine monitoring, for instance, machine breakdown may be very rare and examples of this event may be unavailable. In such cases, the most widely accepted approach is to model the normal behavior of the system in order to subsequently detect unknown events. This is the basic concept of One-Class Classification (OCC). In some preliminary works we have proposed the Feature Boundaries Detector for One-Class Classification FBDOCC method, which operates by examining each problem feature at a time. In this paper we propose an extensive study of the behavior of the FBDOCC and introduce another version of the method, namely FBDOCC2. This work, also considers the use of the Particle Swarm Optimization (PSO) algorithm to find the best parameter configuration of the proposed method. Furthermore, this work also introduces a procedure to improve the training time performance without degrading the quality of the classification as well as some other contributions hereafter described. A number of experiments were carried out with synthetic and real datasets aiming at comparing both versions of the FBDOCC method with the most recent and effective OCC methods, namely: SVDD, One-class SVM, Least Squares One-class SVM, Kernel PCA, Gaussian Process Prior OCC, Condensed Nearest Neighbor Data Description and One-class Random Forests. The evaluation metrics considered in the experiments are: (i) the Area Under the ROC Curve (AUC); (ii) the Matthews Correlation Coefficient (MCC); (iii) the training time; and (iv) the prototype reduction rate. Regarding the metrics AUC and MCC, the first FBDOCC version has presented the best overall results among all the methods whereas the FBDOCC2 obtained results comparable to the best methods in some experiments where the standard FBDOCC yielded a poorer performance. FBDOCC was the faster method to train in comparison to the other methods in all but one dataset. In addition, FBDOCC was much faster than all methods based on support vector machines. [ABSTRACT FROM AUTHOR]

Highlights: [•] Point-feature cartographic label placement problem (PFCLP) arises in design of maps. [•] This paper examines the PFCLP in the legibility context. [•] The PFCLP is solved as dispersion problems based on binary integer linear programming. [ABSTRACT FROM AUTHOR]

Abstract: The internet is a valuable source of information where many ideas can be found dealing with different topics. A few numbers of ideas might be able to solve an existing problem. However, it is time-consuming to identify these ideas within the large amount of textual information in the internet. This paper introduces a new web mining approach that enables an automated identification of new technological ideas extracted from internet sources that are able to solve a given problem. It adapts and combines several existing approaches from literature: approaches that extract new technological ideas from a user given text, approaches that investigate the different idea characteristics in different technical domains, and multi-language web mining approaches. In contrast to previous work, the proposed approach enables the identification of problem solution ideas in the internet considering domain dependencies and language aspects. In a case study, new ideas are identified to solve existing technological problems as occurred in research and development (R&D) projects. This supports the process of research planning and technology development. [Copyright &y& Elsevier]

Due to the actual operation of the wind turbine, the collected fault data sets are limited, and it is difficult to realize fault diagnosis through the correlation between variables. To solve this problem, a fault diagnosis method based on virtual expansion and spherical mapping model is proposed in this paper. Firstly, Hermite interpolation is applied to discrete wind power data samples to obtain an interpolation curve about the characteristics of the sample, and a synchronous sampling method is adopted for the interpolation curve to construct a virtual sample. Then, the features of the virtual sample are mapped to a three-dimensional space. Define the spherical data model and perform spherical fitting in a three-dimensional coordinate system. Finally, feature extraction is performed on the fitted spherical surface for training and testing extreme learning machine (ELM). The distribution law of fault data in the spherical model is summarized. Using the data generated based on the Bootstrap method as a control group, comparative experiments were carried out in back-propagation neural network (BP), Probabilistic Neural Network (PNN), General Regression Neural Network (GRNN), and Support Vector Machine (SVM), which verified the effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]