Search

The rapid development of intelligent transportation systems (ITS) and the emergency of ever-growing vehicular applications pose significant challenges to the underlying communication system. Without the support of powerful communication, many vehicular services will just stay in conceptual phase and cannot be put into practice. Recently, vehicular fog computing (VFC) is introduced as a promising solution to provide low-latency services on roads, which utilizes enormous vehicles on roads as communication and computation resources and extends cloud computing service to an edge network. On the other hand, the publish/subscribe (pub/sub) paradigm provides a loosely coupled and scalable communication which can facilitate flexible and dynamic vehicular network services. Motivated by the merits of these two research fields, in this paper, we propose a novel joint design of pub/sub-communication model based on VFC architecture, which employs fog nodes as the data platform for messages aggregation. Specifically, we describe a method to predict the vehicle’s destination and construct a stable VFC on roads. Then, a message dissemination approach is designed based on our communication model. Finally, the experimental results confirm the efficiency of our proposed scheme in real-world urban scenarios.

This paper developed a multiobjective Big Data optimization approach based on a hybrid salp swarm algorithm and the differential evolution algorithm. The role of the differential evolution algorithm is to enhance the capability of the feature exploitation of the salp swarm algorithm because the operators of the differential evolution algorithm are used as local search operators. In general, the proposed method contains three stages. In the first stage, the population is generated, and the archive is initialized. The second stage updates the solutions using the hybrid salp swarm algorithm and the differential evolution algorithm, and the final stage determines the nondominated solutions and updates the archive. To assess the performance of the proposed approach, a series of experiments were performed. A set of single-objective and multiobjective problems from the 2015 Big Data optimization competition were tested; the dataset contained data with and without noise. The results of our experiments illustrated that the proposed approach outperformed other approaches, including the baseline nondominated sorting genetic algorithm, on all test problems. Moreover, for single-objective problems, the score value of the proposed method was better than that of the traditional multiobjective salp swarm algorithm. When compared with both algorithms, that is, the adaptive DE algorithm with external archive and the hybrid multiobjective firefly algorithm, its score was the largest. In contrast, for the multiobjective functions, the scores of the proposed algorithm were higher than that of the fireworks algorithm framework.

Generating text to have a consistent style with only a few observed highly-stylized text samples is a difficult task for image processing. The text style involving the typography, i.e., font, stroke, color, decoration, effects, etc., should be considered for transfer. In this paper, we propose a novel approach to stylize target text by decoding weighted deep features from only a few referenced samples. The deep features, including content and style features of each referenced text, are extracted from a Convolutional Neural Network (CNN) that is optimized for character recognition. Then, we calculate the similarity scores of the target text and the referenced samples by measuring the distance along the corresponding channels from the content features of the CNN when considering only the content, and assign them as the weights for aggregating the deep features. To enforce the stylized text to be realistic, a discriminative network with adversarial loss is employed. We demonstrate the effectiveness of our network by conducting experiments on three different datasets which have various styles, fonts, languages, etc. Additionally, the coefficients for character style transfer, including the character content, the effect of similarity matrix, the number of referenced characters, the similarity between characters, and performance evaluation by a new protocol are analyzed for better understanding our proposed framework.

In this paper, we present a novel double graph regularized double dictionary learning (DGRDDL) method for image classification. The proposed method jointly constructs a number of class-specific sub-dictionaries to capture the most discriminative features (class-specific information) of each class, and a class-shared dictionary to model the common patterns (class-shared information) shared by the images from different classes. A novel double graph regularization is proposed to correctly represent and differentiate these two types of information. Specifically, an intra-class similarity graph constraint is imposed on the representation coefficients over the class-specific dictionaries, and an inter-class similarity graph constraint is applied on the representation coefficients over the class-shared dictionary. In this way, the representations learned by the proposed DGRDDL method can correctly model the local similarity relationships of the class-specific and the class-shared information in images, respectively. Moreover, due to the differences between the intra-class and inter-class similarity graphs, the two types of information can be appropriately separated and captured by the learned dictionaries. We evaluate the performance of the proposed method on six public datasets and compared against those of seven benchmark methods. The experimental results demonstrate the effectiveness and superiority of the proposed method in image classification over the benchmark dictionary learning methods.

Many undergraduate students majoring in computer science (CS) can not clearly state the complete execution process of a C program in a computer system, although they have learned many CS courses such as programming, assembly and circuits when they graduate. The reason for this phenomenon is because they lack the computer system perspective and cannot integrate the knowledges of different courses. To solve this problem, we opened a new course computer system fundamentals (CSF) aiming to enable the students to understand how each statement of the C program realizes its functionality by assemble codes (ACs), how these ACs are converted into the machine codes (MCs), how the MCs are stored in the memory system, how these MCs are loaded into CPUs, and how the CPUs execute the MCs. This course can help the students integrate the knowledge of programming, assembly, machine code, instruction set, digital logic and circuits, thereby making them establish a full-stack computer system perspective, build the hardware and software co-optimization thinking, and also improve the ability to solve the complex engineering problems. This course has been teaching in the Internet of Things Engineering major for 4 years, and the course survey result proves it has improved the students’ CSA effectively.

Most of the deep learning based speech enhancement methods focus on the modeling of complicated relationship between the noisy speech and the clean speech without the consideration of noise information. In order to cope with various complex noise scenes, we introduce a novel enhancement architecture that integrates a deep autoencoder with neural noise embedding. In this study, a new normalization method, termed conditional layer normalization (CLN), is introduced to improve the generalization of deep learning based speech enhancement approaches for unseen environments. The noise embedding is passed through the CLN layers to regularize the network for speech enhancement task. The proposed network can be adaptively adjusted according to different noise information extracted from the noisy speech input. The network in overall is trained in an end-to-end manner and the experimental results show that the proposed scheme produces satisfactory enhancement performance comparing the other methods. The visualization shows that our proposed network captures noise information, which is helpful to improve robustness to unseen environments for speech enhancement.

Android users are occasionally troubled by the slow UI responses and sudden application/OS crashes. These problems are mainly caused by software aging, a phenomenon of progressive degradation of performance and dependability typically observed in long-running software systems. A countermeasure to software aging is software rejuvenation, i.e., manual or scheduled restart at different levels, such as application, OS, and device. Various software aging and rejuvenation models have been proposed for different software systems. However, these traditional models cannot be applied in the context of mobile devices, as they seldom consider the patterns of usage behavior and user experience specific to mobile phones. We address this problem based on the observations that the usage time of mobile phones is typically fragmented in daily life, with frequent and periodical switches between active and sleep modes, and that the user experience on fluent operation in the active mode is a key concern for mobile users. These insights are exploited to model the usage behavior and aging process by individual Stochastic Petri-Nets, and then to compose them into a Continuous Time Markov Chain (CTMC). Furthermore, we propose proactive rejuvenation strategies based on such CTMCs to achieve the best user experience and the least user interference, such as restarting the device when it is in sleep mode and before it enters an aged state. To consider user experience - a key concern of mobile users which is still less prominent in traditional dependability measurements – we propose new related metrics: for fluency (i.e., the probability that a phone offers a fast UI response to the users), and for the degree of overall user experience. We demonstrate the effectiveness and advantages of the proposed models and metrics via simulations as well as an empirical study.

Image segmentation is among the most important techniques in image processing, and many methods have been developed to perform this task. This paper presents a new multi-objective metaheuristic based on a multi-verse optimization algorithm to segment grayscale images via multi-level thresholding. The proposed approach involves finding an approximate Pareto-optimal set by maximizing the Kapur and Otsu objective functions. Both Kapur’s and Otsu’s methods are highly used for image segmentation performed by means of bi-level and multi-level thresholding. However, each of them has certain characteristics and limitations. Several metaheuristic approaches have been proposed in the literature to separately optimize these objective functions in terms of accuracy, whereas only a few multi-objective approaches have explored the benefits of the joint use of Kapur and Otsu’s methods. However, the computational cost of Kapur and Otsu is high and their accuracy needs to be improved. The proposed method, called Multi-objective Multi-verse Optimization, avoids these limitations. It was tested using 11 natural grayscale images and its performance was compared against three of well-known multi-objective algorithms. The results were analyzed based on two sets of measures, one to assess the performance of the proposed method as a multi-objective algorithm, and the other to evaluate the accuracy of the segmented images. The results showed that the proposed method provides a better approximation to the optimal Pareto Front than the other algorithms in terms of hypervolume and spacing. Moreover, the quality of its segmented image is better than those of the other methods in terms of uniformity measures.

With the advent of social network services, Arabs’ opinions on the web have attracted many researchers in recent years toward detecting and classifying sentiments in Arabic tweets and reviews. However, the impact of word embeddings vectors (WEVs) initialization and dataset balance on Arabic sentiment classification using deep learning has not been thoroughly studied. In this article, a multi-channel embedding convolutional neural network (MCE-CNN) is proposed to improve Arabic sentiment classification by learning sentiment features from different text domains, word, and character n-grams levels. MCE-CNN encodes a combination of different pre-trained word embeddings into the embedding block at each embedding channel and trains these channels in parallel. Besides, a separate feature extraction module implemented in a CNN block is used to extract more relevant sentiment features. These channels and blocks help to start training on high-quality WEVs and fine-tuning them. The performance of MCE-CNN is evaluated on several standard balanced and imbalanced datasets to reflect real-world use cases. Experimental results show that MCE-CNN provides a high classification accuracy and benefits from the second embedding channel on both standard Arabic and dialectal Arabic text, which outperforms state-of-the-art methods.

This paper presents an alternative method for cloud task scheduling problem which aims to minimize makespan that required to schedule a number of tasks on different Virtual Machines (VMs). The proposed method is based on the improvement of the Moth Search Algorithm (MSA) using the Differential Evolution (DE). The MSA simulates the behavior of moths to fly towards the source of light in nature through using two concepts, the phototaxis and Levy flights that represent the exploration and exploitation ability respectively. However, the exploitation ability is still needed to be improved, therefore, the DE can be used as local search method. In order to evaluate the performance of the proposed MSDE algorithm, a set of three experimental series are performed. The first experiment aims to compare the traditional MSA and the proposed algorithm to solve a set of twenty global optimization problems. Meanwhile, in second and third experimental series the performance of the proposed algorithm to solve the cloud task scheduling problem is compared against other heuristic and meta-heuristic algorithms for synthetical and real trace data, respectively. The results of the two experimental series show that the proposed algorithm outperformed other algorithms according to the performance measures.

Automatic image annotation becomes a hot research area because of its efficiency on shrinking the semantic gap between images and their semantic meanings. We present a model referred as weight-KNN which firstly introduces the CNN feature to address the problem that traditional models only work well with well-designed manual feature representations. Additionally, in order to employ the simplicity and generality of the KNN-based model for annotation, the proposed model incorporates a multi-label linear discriminant approach to compute the weighting which improves the accuracy in the subsequent procedures of distance calculation. Moreover, we take the advantage of the KNN-based model to acquire the test image’s k-nearest neighbors in each label category and get the prediction of the image according to the contribution of its neighbors. At last, the experiments are performed on three typical image data sets, corel 5k, esp game and laprtc12, which verify the effectiveness of the proposed model.

In recent years, convolutional neural network (CNN) has attracted considerable attention since its impressive performance in various applications, such as Arabic sentence classification. However, building a powerful CNN for Arabic sentiment classification can be highly complicated and time consuming. In this paper, we address this problem by combining differential evolution (DE) algorithm and CNN, where DE algorithm is used to automatically search the optimal configuration including CNN architecture and network parameters. In order to achieve the goal, five CNN parameters are searched by the DE algorithm which include convolution filter sizes that control the CNN architecture, number of filters per convolution filter size (NFCS), number of neurons in fully connected (FC) layer, initialization mode, and dropout rate. In addition, the effect of the mutation and crossover operators in DE algorithm were investigated. The performance of the proposed framework DE-CNN is evaluated on five Arabic sentiment datasets. Experiments’ results show that DE-CNN has higher accuracy and is less time consuming than the state-of-the-art algorithms.

To prevent the same known vulnerabilities from affecting different firmware, searching known vulnerabilities in binary firmware across different architectures is crucial. Because the accuracy of existing cross-architecture vulnerability search methods is not high, we propose a staged approach based on support vector machine (SVM) and attributed control flow graph (ACFG) at the function level to improve the accuracy using prior knowledge. Furthermore, for efficiency, we utilize the k-nearest neighbor (kNN) algorithm to prune and SVM to refine in the function prefilter stage. Although the accuracy of the proposed method using kNN-SVM approach is slightly lower than the accuracy of the method using only SVM, its efficiency is significantly enhanced. We have implemented our approach CVSkSA to search several vulnerabilities in real-world firmware images. The experimental results show that the accuracy of the proposed method using kNN-SVM approach is close to the accuracy of the method using only SVM in most cases, while the former is approximately four times faster than the latter.

In this paper, a multiscale contour steered region integral (MCSRI) method is proposed to classify highly similar shapes with flexible interior connection architectures. A component distance map (CDM) is developed to robustly characterize the flexible interior connection structure, shape of the exterior contour, and their inter-relationship in a shape image. A novel multiscale region transform (MReT) is proposed to perform region integral over different contour-steered strips at all possible scales to effectively integrate patch features, and thus enables a better description of the shape image in a coarse-to-fine manner. It is applied to solve a challenging problem of classifying cultivars from leaf images, which is a new attempt in both biology and computer vision research communities. A soybean cultivar leaf vein database (SoyCultivarVein), which is the first cultivar leaf vein database, is created and presented for performance evaluation. The experimental results demonstrate the superiority of the proposed method over the state-of-the-art methods in similar shape classification and the possibility of cultivar recognition via leaf pattern analysis, which may lead to a new research interest towards fine-level shape analysis on cultivar classification.

Vision-based localization and mapping in the agricultural environment is challenging due to the unstructured scene with unstable features, illumination variations, bumpy roads, and dynamic environmental objects. To address these challenges, we propose an accurate and robust stereo direct visual odometry system with modifications on Stereo-DSO. We firstly select some well-matched static stereo points in the latest keyframe to improve the accuracy of inverse depth calculation for tracking. The inverse depth can further distinguish close objects from background, which will avoid large and far-away scene objects in keyframe determination. To boost efficiency and accuracy at the tracking stage, we propose a point selection method to sample map points and remove outliers. Furthermore, altitude smoothness verification with a local flat ground assumption and recovery method for tracking failure on bumpy roads are proposed to improve the system’s robustness. Finally, a far-away keyframe is reserved in the sliding window to alleviate the orientation drift since the agricultural robots usually move straightly following the crop row. Our system achieved new state-of-the-art results on Flourish dataset and the recently released Rosario dataset.

Object proposal technique with dense anchoring scheme for scene text detection were applied frequently to achieve high recall. It results in the significant improvement in accuracy but waste of computational searching, regression and classification. In this paper, we propose an anchor selection-based region proposal network (AS-RPN) using effective selected anchors instead of dense anchors to extract text proposals. The center, scales, aspect ratios and orientations of anchors are learnable instead of fixing, which leads to high recall and greatly reduced numbers of anchors. By replacing the anchor-based RPN in Faster RCNN, the AS-RPN-based Faster RCNN can achieve comparable performance with previous state-of-the-art text detecting approaches on standard benchmarks, including COCO-Text, ICDAR2013, ICDAR2015 and MSRA-TD500 when using single-scale and single model (ResNet50) testing only.

Face anti-spoofing is becoming increasingly indispensable for face recognition systems, which are vulnerable to various spoofing attacks performed using fake photos and videos. In this paper, a novel “LDN-TOP representation followed by ProCRC classification” pipeline for face anti-spoofing is proposed. We use local directional number pattern (LDN) with the derivative-Gaussian mask to capture detailed appearance information resisting illumination variations and noises, which can influence the texture pattern distribution. To further capture motion information, we extend LDN to a spatial-temporal variant named local directional number pattern from three orthogonal planes (LDN- TOP). The multi-scale LDN- TOP capturing complete information is extracted from color images to generate the feature vector with powerful representation capacity. Finally, the feature vector is fed into the probabilistic collaborative representation based classifier (ProCRC) for face anti-spoofing. Our method is evaluated on three challenging public datasets, namely CASIA FASD, Replay-Attack database, and UVAD database using sequence-based evaluation protocol. The experimental results show that our method can achieve promising performance with 0.37% EER on CASIA and 5.73% HTER on UVAD. The performance on Replay-Attack database is also competitive.

As a cross modality retrieval task, RGB-infrared person re-identification(Re-ID) is an important and challenging task, because of its important role in video surveillance applications and large cross-modality variations between visible and infrared images. Most previous works addressed the problem of cross-modality gap with feature alignment by original feature representation learning straightly. In this paper, different from existing works, we propose a novel network(CE2L) to tackle the cross-modality gap with feature alignment. CE2L mainly focuses on adding discriminative information and learning robust features by converting modality between visible and infrared images. Its merits are highlighted in two aspects: 1)Using CycleGAN to convert infrared images into color images can not only increase the recognition characteristics of images, but also allow the network to better learn the two modal image features; 2)Our novel method can serve as data augmentation. Specifically, it can increase data diversity and total data against over-fitting by converting labeled training images to another modal images. Extensive experimental results on two datasets demonstrate superior performance compared to the baseline and the state-of-the-art methods.

Knowledge graph reasoning aims at solving certain tasks by finding reasoning paths, which has aroused extensive attention. Recently, a solution for path reasoning that combines reinforcement learning has achieved successful progress. But these researches mainly focus on the agent’s choice of relation and ignore the importance of entity, which will cause the random selection by the agent if 1-N/N-N relations occur. Thus, we propose a reinforcement learning based path reasoning model, which solves this problem from the topological and semantic levels. First, the attention mechanism is introduced in our model, which can extract the hidden feature from neighbor entities and helps the policy network to make a suitable choice instead of random for the actions with the same relation. Then, we introduce a convolutional neural network into our model to distinguish the rationality of the path by the semantic feature. To mitigate the negative impact of terminal rewards, we use a potential-based reward shaping function, which considers the potential gap between agent states as the reward and without any pre-training. Finally, we compare our model with the state-of-the-art baselines on two benchmark datasets, the results of extensive comparison experiments validate the effectiveness of the proposed method.

Few-shot learning aims to train a classifier which can recognize a new class from a few examples like a human. Recently, some works have leveraged auxiliary information in few-shot learning, such as textual data, unlabeled visual data. But these data are positive data, they are close related to the training data. Different from such experimental settings, people can also get knowledge from negative data to better recognize a new class. Inspired by this, we exposure a few unlabeled outliers in each few-shot learning tasks to assist the learning of the classifier. To the best of our knowledge, we are the first ones who propose to utilize outliers to improve few-shot learning. We propose a novel method based on meta-learning paradigms to utilize unlabeled outliers. We not only utilize unlabeled outliers to optimize the meta-embedding network but also adaptively leverage them to enhance the class prototypes. Experiments show that our outlier exposure network can improve few-shot learning performance with a few unlabeled outliers exposure.

Facial makeup transfer aims to render a non-makeup face image in an arbitrary given makeup one while preserving face identity. The most advanced method separates makeup style information from face images to realize makeup transfer. However, makeup style includes several semantic clear local styles which are still entangled together. In this paper, we propose a novel unified adversarial disentangling network to further decompose face images into four independent components, i.e., personal identity, lips makeup style, eyes makeup style and face makeup style. Owing to the disentangled makeup representation, our method can not only flexible control the degree of local makeup styles, but also can transfer local makeup styles from different images into the final result, which any other approaches fail to handle. For makeup removal, different from other methods which regard makeup removal as the reverse process of makeup transfer, we integrate the makeup transfer with the makeup removal into one uniform framework and obtain multiple makeup removal results. Extensive experiments have demonstrated that our approach can produce visually pleasant and accurate makeup transfer results compared to the state-of-the-art methods.

In view of the lack of labeled corpus and the difficulty of extracting multiple relations in the extraction of entity relation in the geographic domain, a method based on probabilistic graph is proposed for extracting the entity relation in the geographic domain. This method uses the semantic information in the knowledge base to enhance the representation of the geographic domain corpus to alleviate the problem of insufficient labeled corpus. It uses character-word hybrid vectors that can be effectively integrated into the semantic information as the feature vectors. The vectors are transmitted to Bi-LSTM and self-attention for global deep feature extraction. Finally drawing on the idea of probabilistic graph, the "semi pointer-semi annotation" method is utilized to extract the head entities, traverse the head entity, and then uses the same method to extract tail entities and relations. By comparing the experimental results on the geographic domain corpus and ACE05 corpus with other advanced methods, the probabilistic graph-based extraction method effectively improves the geographic domain entity relation extraction effect.

Wheat is one of the main food crops in the world today. In order to increase wheat yield, breeding experts are committed to discovering the connection between its genotype and phenotype. Existing phenotype extraction methods mostly rely on manual methods, and the amount of collected data is limited and inefficient. 3D Computed Tomography (CT) imaging has the advantages of fine imaging, high dynamic range, and nondestructive detection of internal structures, which can help to extract the high-throughput phenotype of wheat non-destructively and finely. However, the 3D images scanned by CT have the characteristics of large amount of data and highly sparse content, which brings great challenges to extract phenotype from them. We propose an adaptive supervoxel algorithm for the high sparsity of 3D CT images, which is used for the pre-segmentation. Then, an octree convolutional network is used to optimize the sparse structure storage, and further improve the efficiency of fine segmentation. At last, phenotype is extracted automatically from the finely segmented wheat grain.

Petrol engine becomes much smaller and has higher compression ratio due to the strict exhaust emission regulations and consumer demand for power performance. Meanwhile, the frequency of super-knock is increasing sharply at low speed of high load working region, which has great damage to engine components. Prediction of super-knock is therefore of great significance to optimize operation conditions and prolong service life. High-performance in-cylinder sensors can provide quality information for super-knock prediction, but at a price that production vehicles cannot afford. An approach that does not require additional engine sensors while having an acceptable accuracy rate is needed. Furthermore, the super-knock events are difficult to capture, making prediction task subjects to sample imbalance. In this research, a deep learning approach using only existing engine sensors was applied to address the challenge. Specifically, a two-stage model was designed: (i) Feature extraction, an LSTM based triplet network is built to extract the differential features between super-knock and normal combustion from the engine working cycles. (ii) Classification, to effectively use the extracted features and metric information from the previous stage, a modified KNN classifier is employed. The effectiveness of the proposed approach, i.e., TLSTM-KNN was evaluated and compared with other state-of-the-art methods on an actual engine-based dataset under different imbalance ratios (IR). The prediction precision and recall of super-knock event reach to 81.67% and 79.01% for the test dataset at IR=10. Even at IR=100, the precision and recall reach to 59.02% and 63.31%. The comparisons suggested that the proposed model provides better prediction performance and robustness capability compared to other traditional prediction models. Finally sensitivity analysis was used to explore the factors influencing super-knock.

Vehicular communication networking (VCN) has been widely recognized as a promising solution to support safety-related applications in urban transportation systems. In VCN, efficient message dissemination can let vehicles be better aware of the potential risks and traffic anomalies, which is critical to road safety and traffic efficiency. Substantial studies have focused on the design of inter-vehicle message dissemination protocols. Nonetheless, most existing designs only consider the rapid end-to-end transmission, few of which take into account the broadcast coverage. In this paper, we propose a new message dissemination scheme in the urban traffic scenario by considering both the time constraint and the spatial distribution of data dissemination. Specifically, based on the temporal and spatial correlation of vehicle trajectory, relay vehicles are selected to construct a temporary warning network (TWN) for a rapid safety message dissemination in the regions of interest (ROI). Finally, we conduct extensive numerical experiments to validate the effectiveness of our method in various traffic scenarios.

This paper proposes a novel approach for pose transfer, which aims at transferring the pose of a given person to a target pose. Unlike previous works directly simulating the target pose, we emphasize the pose geometric constraint in our approach to tackle this task. To capture the geometric constraints among the condition pose and target pose, we generate a sequence of different intermediate pose skeleton. Moreover, we introduced the 3D convolution into the generator of our Generative Adversarial Network (GAN) in order to learn spatiotemporal features from the generated pose skeleton sequence. The final target pose will be estimated and refined by a pose transfer block and a series of residual blocks based on the spatiotemporal features. Compared with the existing works, our generated person pose images achieved better performance in terms of Inception Score and Structure Similarity. Extensive experiments on dataset Market-1501 demonstrate the effectiveness of the pose skeleton sequence on pose transfer.

Vehicular ad hoc networks (VANETs) have been widely recognized as a promising solution to improve traffic safety and efficiency for the ability to provide situation awareness even though the potential dangers and traffic anomalies are out of the visual range. In VANETs, time-division multiple access (TDMA) based overlay protocols can prevent transmission collisions, and play an important role in providing an efficient communication channel. However, due to high vehicle mobility and time-varying traffic flow, the existing TDMA-based slot allocation approaches cannot fully utilize the channel resources, which may result in high transmission delay and packet collision. To overcome these shortcomings, we propose a traffic-aware TDMA-based MAC (TA-MAC) protocol which utilizes the capability of mobile edge computing (MEC) in this paper. Specifically, based on MEC and vehicle-to-road-side-units (V2R) communications, a traffic-aware mechanism is first proposed to estimate the traffic condition on the road segment. Then, we propose a new slot assignment method that aims at guaranteeing the high channel utilization and low delay of safety message under dynamic traffic conditions. Finally, we conduct extensive experiments to demonstrate the effectiveness of the proposed protocol.

Prototypical networks have shown their simplicity and effectiveness in few-shot learning. Recently, some works leverage cross-modal information to enhance the class prototype in few-shot learning. However, they seldom make use of structured information from textual space to optimize class prototype representation. And they either don’t align textual modality to visual modality or align them too rigidly. We argue that proper alignment method is important to improve the performance of cross-modal methods, as query data only has visual information in the few-shot learning tasks. In this paper, we propose a cross-modal alignment method to optimize class prototypes with structured information from textual space. We further introduce an anisotropic combination method to enhance class prototypes with information from two modalities. Experiments show that the cross-modal alignment method and anisotropic combination method achieve state-of-the-art results on the miniImageNet and tieredImageNet benchmark in one-shot regime.

Text categorization or Textual document is a method that becomes more significant in tagging a textual document to their most relevant label. However, not all languages have parallel textual growth; without free and absences of a dataset, text categorization becomes interesting for Tigrigna language, i.e., low-resourced language. Our aim to identify the given document to its categories based on its linguistic features. To achieve our goal, we have constructed a new dataset from different Tigrigna news sources. The dataset has six main categories: Agriculture, Sports, Health, Education, Religion, and Politics. Each collected is article preprocessed from Latin characters, punctuations, and stop words. We deployed a collection of different classical machine learning classifiers to investigate its effectiveness in our datasets. Namely, 7 popular classifiers were used, Logistic Regression, Nearest Centroid, Decision Tree (DT), Support Vector Machines (SVM), K-nearest neighbors (KNN), Random Forest Classifier, and Multi-Layer Perceptron (MLP). Ensemble models also implemented to get the best accuracy by combining the best classifiers based on their majority-voting classifiers. Our experimental results showed reliable performance with a minimum F1-score of 89.1% achieved by Nearest centroid and top performance of 96 % achieved by SVM. The experimental results presented in terms of precision, Recall, and F1-scores.

In recent years, cloud computing technology has attracted extensive attention from both academia and industry. The popularity of cloud computing was originated from its ability to deliver global IT services such as core infrastructure, platforms, and applications to cloud customers over the web. Furthermore, it promises on-demand services with new forms of the pricing package. However, cloud job scheduling is still NP-complete and became more complicated due to some factors such as resource dynamicity and on-demand consumer application requirements. To fill this gap, this paper presents a modified Harris hawks optimization (HHO) algorithm based on the simulated annealing (SA) for scheduling jobs in the cloud environment. In the proposed HHOSA approach, SA is employed as a local search algorithm to improve the rate of convergence and quality of solution generated by the standard HHO algorithm. The performance of the HHOSA method is compared with that of state-of-the-art job scheduling algorithms, by having them all implemented on the CloudSim toolkit. Both standard and synthetic workloads are employed to analyze the performance of the proposed HHOSA algorithm. The obtained results demonstrate that HHOSA can achieve significant reductions in makespan of the job scheduling problem as compared to the standard HHO and other existing scheduling algorithms. Moreover, it converges faster when the search space becomes larger which makes it appropriate for large-scale scheduling problems.

Few-shot classification aims to recognize unlabeled samples from unseen classes given only a small number of labeled examples. Most methods addressing few-shot problem through meta-learning. They focus on learning a generic classifier across a large number of multiclass classification tasks and generalizing the model to a new task. However, the low-data problem in the novel classification task still remains. In this paper, we propose Transductive Data Clustering Transformation (TDCT), a novel and simple method which can potentially be applied to any metric-based few-shot classification approaches. TDCT exploits the task-specific knowledge and enhances the data representations by using a transformation that incorporates data clustering. This transformation implicitly does transductive inference by leveraging the relationships between all samples within a task, alleviating the low data problem. Extensive experiments show that TDCT is an effective and computationally efficient method which can improve few-shot learning performance by a large margin on two benchmarks.

In this paper, we propose an approach to tackle the problem of the automatic restoration of Arabic diacritics that includes three components stacked in a pipeline: a deep learning model which is a multi-layer recurrent neural network with LSTM and Dense layers, a character-level rule-based corrector which applies deterministic operations to prevent some errors, and a word-level statistical corrector which uses the context and the distance information to fix some diacritization issues. This approach is novel in a way that combines methods of different types and adds edit distance based corrections.

Exterior contour and interior structure are both vital features for classifying objects. However, most of the existing methods consider exterior contour feature and internal structure feature separately, and thus fail to function when classifying patchy image structures that have similar contours and flexible structures. To address above limitations, this paper proposes a novel Multi-Orientation Region Transform (MORT), which can effectively characterize both contour and structure features simultaneously, for patchy image structure classification. MORT is performed over multiple orientation regions at multiple scales to effectively integrate patchy features, and thus enables a better description of the shape in a coarse-to-fine manner. Moreover, the proposed MORT can be extended to combine with the deep convolutional neural network techniques, for further enhancement of classification accuracy. Very encouraging experimental results on the challenging ultra-fine-grained cultivar recognition task, insect wing recognition task, and large variation butterfly recognition task are obtained, which demonstrate the effectiveness and superiority of the proposed MORT over the state-of-the-art methods in classifying patchy image structures. Our code and three patchy image structure datasets are available at: https://github.com/XiaohanYu-GU/MReT2019., Accepted by AAAI 2020

Motivation The fundamental challenge of modern genetic analysis is to establish gene-phenotype correlations that are often found in the large-scale publications. Because lexical features of gene are relatively regular in text, the main challenge of these relation extraction is phenotype recognition. Due to phenotypic descriptions are often study- or author-specific, few lexicon can be used to effectively identify the entire phenotypic expressions in text, especially for plants. Results We have proposed a pipeline for extracting phenotype, gene and their relations from biomedical literature. Combined with abbreviation revision and sentence template extraction, we improved the unsupervised word-embedding-to-sentence-embedding cascaded approach as representation learning to recognize the various broad phenotypic information in literature. In addition, the dictionary- and rule-based method was applied for gene recognition. Finally, we integrated one of famous information extraction system OLLIE to identify gene-phenotype relations. To demonstrate the applicability of the pipeline, we established two types of comparison experiment using model organism Arabidopsis thaliana. In the comparison of state-of-the-art baselines, our approach obtained the best performance (F1-Measure of 66.83%). We also applied the pipeline to 481 full-articles from TAIR gene-phenotype manual relationship dataset to prove the validity. The results showed that our proposed pipeline can cover 70.94% of the original dataset and add 373 new relations to expand it. Availability and implementation The source code is available at http://www.wutbiolab.cn: 82/Gene-Phenotype-Relation-Extraction-Pipeline.zip. Supplementary information Supplementary data are available at Bioinformatics online.

Biometrics have been widely studied in recent years, and they are increasingly employed in real-world applications. Meanwhile, a number of potential threats to the privacy of biometric data arise. Iris template protection demands that the privacy of iris data should be protected when performing iris recognition. According to the international standard ISO/IEC 24745, iris template protection should satisfy the irreversibility, revocability, and unlinkability. However, existing works about iris template protection demonstrate that it is difficult to satisfy the three privacy requirements simultaneously while supporting effective iris recognition. In this paper, we propose an iris template protection method based on local ranking. Specifically, the iris data are first XORed (Exclusive OR operation) with an application-specific string; next, we divide the results into blocks and then partition the blocks into groups. The blocks in each group are ranked according to their decimal values, and original blocks are transformed to their rank values for storage. We also extend the basic method to support the shifting strategy and masking strategy, which are two important strategies for iris recognition. We demonstrate that the proposed method satisfies the irreversibility, revocability, and unlinkability. Experimental results on typical iris datasets (i.e., CASIA-IrisV3-Interval, CASIA-IrisV4-Lamp, UBIRIS-V1-S1, and MMU-V1) show that the proposed method could maintain the recognition performance while protecting the privacy of iris data.

In cognitive radio ad hoc networks (CRAHNs), the secondary users' links are interrupted by the arrival of primary users, which leads to a significant increase of the number of transmissions per packet. Since network coding and opportunistic routing can reduce the number of transmissions over unreliable wireless links, they are more suitable for CRAHNs. Network coding-based opportunistic routing has been well studied in previous works for traditional ad hoc networks. However, existing approaches have limitations to handle CRAHNs. First, the routing decisions of these methods aim to find one optimal forwarding set for the whole path from the source to the destination. If the channel conditions change dynamically, the pre-selected forwarding set may become unavailable or non-optimal. Second, their coding schemes are not efficient enough to cope with the dynamic spectrum variation in CRAHNs. In this paper, we attempt to overcome these limitations and propose a network-coding-based geographic segmented opportunistic routing scheme for CRAHNs, by fully embracing its characteristics. In our scheme, the whole path from the source to the destination is cut down into several smaller opportunistic route segments, where the packets are transmitted through multiple segments based on a step-by-step forwarding procedure until all the packets are delivered to the destination. Since our scheme utilizes only local spectrum opportunities, topology information, and geometric conditions to compute the forwarding set for each short-term opportunistic route segment, it can better adapt to dynamic spectrum environments and changing network topologies in CRAHNs. Furthermore, we construct a coding graph to show that our coding problem is a reduction from the maximum clique problem, and we propose an efficient network coding strategy to solve it. Simulation results show that our scheme can achieve a considerable performance, compared with the latest opportunistic routing protocols designed for CRAHNs.

In this paper, we propose a novel low-rank double dictionary learning (LRD2L) method for robust image classification tasks, in which the training and testing samples are both corrupted. Unlike traditional dictionary learning methods, LRD2L simultaneously learns three components from corrupted training data: 1) a low-rank class-specific sub-dictionary for each class to capture the most discriminative class-specific features of each class, 2) a low-rank class-shared dictionary which models the common patterns shared in the data of different classes, and 3) a sparse error term to model the noise in data. Through low-rank class-shared dictionary and noise term, the proposed method can effectively separate the corruptions and noise in training samples from creating low-rank class-specific sub-dictionaries, which are employed for correctly reconstructing and classifying testing images. Comparative experiments are conducted on three public available databases. Experimental results are encouraging, demonstrating the effectiveness of the proposed method and its superiority in performance over the state-of-the-art dictionary learning methods.

Ultra-fine-grained visual categorization (ultra-FGVC) categorizes objects with more similar patterns between classes than those in fine-grained visual categorization (FGVC), e.g., where the spectrum of granularity significantly moves down from classifying species to classifying cultivars within the same species. It is considered as an open research problem mainly due to the following challenges. First, the inter-class differences among images are much smaller by level of orders (e.g., cultivars in the same species) than those in current FGVC tasks (e.g., species). Second, there is only a few samples per category, which is beyond the ability of most large training data favored convolutional neural network methods. To address these problems, we propose a novel random mask covariance network (MaskCOV), which integrates an auxiliary self-supervised learning module with a powerful in-image data augmentation scheme for the ultra-FGVC. Specifically, we first uniformly partition input images into patches and then augment data by randomly shuffling and masking these patches. On top of that, we introduce an auxiliary self-supervised learning module of predicting the spatial covariance context of these patches to increase discriminability of our network for classification. Very encouraging experimental results of the proposed method in comparison with the state-of-the-art benchmarks demonstrate its superiority and potential of MaskCOV concept, which pushes research boundary forward from the fine-grained to the ultra-fine-grained visual categorization.

Driving performance evaluations can contribute to fleet management and lead to safer and more economical driving conditions for manned or driverless fleet vehicles. One approach to driving performance evaluation involves quantitative mapping or categorical labeling of skill levels and categorizing of driving patterns from extraordinarily mild to the most aggressive. This paper presents a big data system for driving performance evaluations of drivers and trips using a probabilistic framework. The proposed framework combines a feature mixture model for scoring driving performance through defined objective comparison criteria and a latent style mixture model for classifying drivers by the main driving styles they exhibit. To demonstrate the effectiveness of the proposed models, we perform both quantitative and qualitative experiments. The results show that the former produces an interpretable and normal scorecard model, while the latter helps build an improved clustering model that represents enhanced driver behavior.

Person retrieval is an important vision task, aiming at matching the images of the same person under various camera views. The key challenge of person retrieval lies in the large intra-class variations among the person images. Therefore, how to learn discriminative feature representations becomes the core problem. In this paper, we propose a deep part-aware representation learning method for person retrieval. First, an improved triplet loss is introduced such that the global feature representations from the same identity are closely clustered. Meanwhile, a localization branch is proposed to automatically localize those discriminative person-wise parts or regions, only using identity labels in a weakly supervised manner. Via the learning simultaneously guided by the global branch and the localization branch, the proposed method can further improve the performance for person retrieval. Through an extensive set of ablation studies, we verify that the localization branch and the improved triplet loss each contributes to the performance boosts of the proposed method. Our model obtains superior (or comparable) performance compared to state-of-the-art methods for person retrieval on the four public person retrieval datasets. On the CUHK03-labeled dataset, for instance, the performance increases from 73.0% mAP and 77.9% rank-1 accuracy to 80.8% (+7.8%) mAP and 83.9% (+6.0%) rank-1 accuracy.

Classification algorithms have shown exceptional prediction results in the supervised learning area. These classification algorithms are not always efficient when it comes to real-life datasets due to class distributions. As a result, datasets for real-life applications are generally imbalanced. Several methods have been proposed to solve the problem of class imbalance. In this paper, we propose a hybrid method combining the preprocessing techniques and those of ensemble learning. The original training set is undersampled by evaluating the samples by stochastic measurement (SM) and then training these samples selected by Multilayer Perceptron to return a balanced training set. The MLPUS (Multilayer perceptron undersampling) balanced training set is aggregated using the bagging ensemble method. We applied our method to the real-life Niger_Rice dataset and forty-four other imbalanced datasets from the KEEL repository in this study. We also compared our method with six other existing methods in the literature, such as the MLP classifier on the original imbalance dataset, MLPUS, UnderBagging (combining random under-sampling and bagging), RUSBoost, SMOTEBagging (Synthetic Minority Oversampling Technique and bagging), SMOTEBoost. The results show that our method is competitive compared to other methods. The Niger_Rice real-life dataset results are 75.6, 0.73, 0.76, and 0.86, respectively, for accuracy, F-measure, G-mean, and ROC with our proposed method. In contrast, the MLP classifier on the original imbalance Niger_Rice dataset gives results 72.44, 0.82, 0.59, and 0.76 respectively for accuracy, F-measure, G-mean, and ROC.

Purpose This work aims to report the effect of different inlet air velocity settings, and the position of the inlet areas under fire-induced conditions in stairwells. Design/methodology/approach The results are evaluated both experimentally and numerically. In sequence, the study also describes the results of three tests under different ventilation conditions, which were performed by simulating a fire of 0.9 m diameter in a stairwell access door. Detailed transient measurements of air temperature in the walls, air velocity settings at the ventilation outlet, smoke temperature through the fans and pressure through the exhaust fans were recorded for the study. Findings These data could be used as benchmark for future numerical validation studies. Based on computational fluid dynamics, the tests were performed using fire dynamics simulator codes, to compare the results of tests and simulations. Originality/value The numerical study was performed in confined floors close to the test areas of a typical stairwell, with different vents and a source of fire. The results show that the lack of symmetry in the air vents settings has greater influence on the plume than the total area of the air inlet.

Software development is a knowledge intensive and a collaborative activity, which mainly depends on knowledge and experience of the software developers. Knowledge management in software industries aims to create an environment for continuous knowledge sharing and creation to remain competitive. This study investigates the current status of knowledge management in three software industries in Sri Lanka on the basis that they claimed to apply knowledge management practices within their organizations using both quantitative and qualitative methods. The knowledge synthesis theory is used in this study to synthesize a variety of knowledge, unseen problems, obstructions, and the corresponding underlying reasons in implementing the knowledge management practices. Findings of this study emphasized that while knowledge management practices were considered to be important, organizational learning and culture, attitudes and behaviors of employees, information technology infrastructure, and rewards systems were inadequate to address an effective management of knowledge in these three software industries. One of the findings showed that, software developers believe both software engineering knowledge and information communication technology are considered to be an enterprise’s strategic asset and a main source to create competitive advantage. Findings also presented a compelling evidence of the soundness for the need of knowledge management practices to manage software engineering knowledge.

Pedestrian navigation at night should differ from daytime navigation due to the psychological safety needs of pedestrians. For example, pedestrians may prefer better-illuminated walking environments, shorter travel distances, and greater numbers of pedestrian companions. Route selection at night is therefore a multi-objective optimization problem. However, multi-objective optimization problems are commonly solved by combining multiple objectives into a single weighted-sum objective function. This study extends the artificial bee colony (ABC) algorithm by modifying several strategies, including the representation of the solutions, the limited neighborhood search, and the Pareto front approximation method. The extended algorithm can be used to generate an optimal route set for pedestrians at night that considers travel distance, the illumination of the walking environment, and the number of pedestrian companions. We compare the proposed algorithm with the well-known Dijkstra shortest-path algorithm ...

The automatic congestion detection of campus traffic presents a significant challenge to the traffic congestion research community. Typically, campus road users can be classified into four types including pedestrian, bike, vehicle and motorbike, which enhances the complexity of traffic condition. Thus, existing descriptors of traffic congestion for highway traffic are not valid when describing the traffic congestion in campus. In this paper, we propose a novel descriptor, road occupancy rate, for measuring campus traffic congestion level, which is statistically proved to be the most effective descriptor among other descriptors (including speed of pedestrian, vehicle, motorbike and bike). Two existing models Markov model and back propagation neural network (BPNN) are introduced in this paper to detect the campus traffic congestion combined with the proposed descriptors. And three phases are defined based on three-phase traffic theory to describe the campus traffic congestion levels. Experimental results indicate that the proposed detecting methods are both capable of detecting campus traffic congestion, while the BPNN-based method achieves higher accuracy and more stable performance.

In this paper, we evaluate the impact of climate parameters on the yield of rice production at the Office du Niger (ON) in Mali using four classifiers. Three machine learning classification algorithms, namely k-Nearest Neighbor (KNN), Support Vector Classifier (SVC), and Logistic Regression (LR) and Multilayer Perceptron (MLP) classifier are utilized. Our experience reveals that the three classification algorithms of machine learning and Multilayer perceptron classifiers give poor results on the original dataset due to the problem of unbalanced data. However, oversampling techniques are applied to the dataset to improve these results. The technique we used is an oversampling technique step by step. Therefore, 100% SMOTE (Synthetic Minority Over-Sampling Technique) is applied to the minority class at each stage, i.e., the minority class is doubled at each step. We found out that the proposed technique gives better results for all these classification algorithms rather than applying SMOTE directly to balance the dataset and other balanced techniques such as class balance and resample. However, MLP gives better results compared to other machine learning classification algorithms using our technique.

This paper presents a novel shape descriptor to effectively and efficiently characterize the local image statistics. The proposed descriptor, termed contour covariance (CC), characterizes covariance features driven by a moving point on the shape contour at multiple scales. To calculate the covariance matrices, three basic features including texture, intensity and distance map, are extracted from the object image. Based on coefficients of the obtained covariance matrices, the proposed CC descriptor is compact yet informative, as well as invariant to rotation, translation and scale. The experimental results on two databases demonstrate the superiority and efficiency of the proposed method among the state-of-the-art methods for shape classification.

Arbitrary characters synthesis based on a few referenced examples poses a great challenge due to the diversity of characters category and style. We regard this problem as image translation problem and propose a character style transfer network consisting of content selector, style encoder, content encoder, feature embedding and embedded feature decoder to solve it. The content selector is used to select and match the most similar content (i.e., font) from our collected glyph dataset as content references. Then, we apply the style encoder and content encoder to extract the style and content representation separately and mix them for feature embedding. Finally, the embedded features are decoded to generate the target characters. We train them in an end-to-end manner and evaluate the proposed method on MC-GAN dataset and our collected dataset. The experimental results have demonstrated the effectiveness of the proposed model for character synthesis.

Existing studies have uncovered that there exist significant Raw Bit Error Rates (RBERs) variations among different layers of 3D flash memories due to manufacture process variation. These RBER variations would cause significantly diversed read latencies when reading data with traditional Low-Density Parity-Check (LDPC) codes designed for planar flash memories, which induces sub-optimal read performance of flash-based Solid-State Drives (SSDs). To investigate the latency diversity, this paper first performs a preliminary experiment and observes that LDPC read levels proportional to latencies increase in diverse speeds along with data retention. Then, by exploiting the observation results, a Multi-Granularity LDPC (MG-LDPC) read method is proposed to adapt level increase speed for each layer. Five LDPC engines with varied increase granularity are designed to adapt RBER speed requirements. Finally, two implementations for MG-LDPC are applied to assign LDPC engines for each flash layer in a fixed way or dynamically according to prior read levels. Experimental results show that the proposed two implementations can reduce SSD read response time by 21% and 47% on average, respectively.CCS CONCEPTS• Hardware→Temperature optimization; Network on chip; 3D integrated circuits.

The extraction of vein traits from venation networks is of great significance to the development of a variety of research fields, such as evolutionary biology. However, traditional studies normally target to the extraction of reticulate structure traits (ReSTs), which is not sufficient enough to distinguish the difference between vein orders. For hierarchical structure traits (HiSTs), only a few tools have made attempts with human assistance, and obviously are not practical for large-scale traits extraction. Thus, there is a necessity to develop the method of automated vein hierarchy classification, raising a new challenge yet to be addressed. We propose a novel vein hierarchy classification method based on directional morphological filtering to automatically classify vein orders. Different from traditional methods, our method classify vein orders from highly dense venation networks for the extraction of traits with ecological significance. To the best of our knowledge, this is the first attempt to automatically classify vein hierarchy. To evaluate the performance of our method, we prepare a soybean transmission image dataset (STID) composed of 1200 soybean leaf images and the vein orders of these leaves are manually coarsely annotated by experts as ground truth. We apply our method to classify vein orders of each leaf in the dataset. Compared with ground truth, the proposed method achieves great performance, while the average deviation on major vein is less than 5 pixels and the average completeness on second-order veins reaches 54.28%.