Abstract Data scarcity is a major challenge when training deep learning (DL) models. DL demands a large amount of data to achieve exceptional performance. Unfortunately, many applications have small or inadequate data to train DL frameworks. Usually, manual labeling is needed to provide labeled data, which typically involves human annotators with a vast background of knowledge. This annotation process is costly, time-consuming, and error-prone. Usually, every DL framework is fed by a significant amount of labeled data to automatically learn representations. Ultimately, a larger amount of data would generate a better DL model and its performance is also application dependent. This issue is the main barrier for many applications dismissing the use of DL. Having sufficient data is the first step toward any successful and trustworthy DL application. This paper presents a holistic survey on state-of-the-art techniques to deal with training DL models to overcome three challenges including small, imbalanced datasets, and lack of generalization. This survey starts by listing the learning techniques. Next, the types of DL architectures are introduced. After that, state-of-the-art solutions to address the issue of lack of training data are listed, such as Transfer Learning (TL), Self-Supervised Learning (SSL), Generative Adversarial Networks (GANs), Model Architecture (MA), Physics-Informed Neural Network (PINN), and Deep Synthetic Minority Oversampling Technique (DeepSMOTE). Then, these solutions were followed by some related tips about data acquisition needed prior to training purposes, as well as recommendations for ensuring the trustworthiness of the training dataset. The survey ends with a list of applications that suffer from data scarcity, several alternatives are proposed in order to generate more data in each application including Electromagnetic Imaging (EMI), Civil Structural Health Monitoring, Medical imaging, Meteorology, Wireless Communications, Fluid Mechanics, Microelectromechanical system, and Cybersecurity. To the best of the authors’ knowledge, this is the first review that offers a comprehensive overview on strategies to tackle data scarcity in DL. 

/pdf/a-survey-on-deep-learning-tools-dealing-with-data-scarcity-3amzf9cp.pdf

A survey on deep learning tools dealing with data scarcity: definitions, challenges, solutions, tips, and applications

The development of the automobile industry and civilian infrastructures improved the lifestyles of everyone in the world. In parallel to the rise in quality of life of everyone, the number of road accidents also rose. The major reason behind road accidents is emotional factors of the drivers. The emotional imbalance will influence the drivers to abandon the traffic rules, neglect speed limits, cross the signals, cross the lane, etc. Recently automobile industries have extended their researches to the development of emotion sensing systems and embedding them inside the vehicles using affective computing technology to mitigate the road accidents. These emotion sensing systems will be decisive and act as human-like driver-assistive systems in alarming the drivers. This chapter focuses on bringing out the feasibility and existing challenges of affective computing in sensing the emotional factors of drivers for improved road safety. 

Feasibility and Necessity of Affective Computing in Emotion Sensing of Drivers for Improved Road Safety

Driving event detection and driver behavior recognition have been widely explored for many purposes, including detecting distractions, classifying driver actions, detecting kidnappings, pricing vehicle insurance, evaluating eco-driving, and managing shared and leased vehicles. Some systems can recognize the main driving events (e.g., accelerating, braking, and turning) by using in-vehicle devices, such as inertial measurement unit (IMU) sensors. In general, feature extraction is a commonly used technique to obtain robust and meaningful information from the sensor signals to guarantee the effectiveness of the subsequent classification algorithm. However, a general assessment of deep neural networks merits further investigation, particularly regarding end-to-end models based on Convolutional Neural Networks (CNNs), which combine two components, namely feature extraction and the classification parts. This paper primarily explores supervised deep-learning models based on 1D and 2D CNNs to classify driving events from the signals of linear acceleration and angular velocity obtained with the IMU sensors of a smartphone placed in the instrument panel of the vehicle. Aggressive and non-aggressive behaviors can be recognized by monitoring driving events, such as accelerating, braking, lane changing, and turning. The experimental results obtained are promising since the best classification model achieved accuracy values of up to 82.40%, and macro- and micro-average F1 scores, respectively, equal to 75.36% and 82.40%, thus, demonstrating high performance in the classification of driving events.

/pdf/evaluation-of-1d-and-2d-deep-convolutional-neural-networks-tiutte3p.pdf

Evaluation of 1D and 2D Deep Convolutional Neural Networks for Driving Event Recognition

In the foreseeable future, electric vehicles (EVs) will play a key role in the decarbonization of transport systems. Replacing vehicles powered by internal combustion engines (ICEs) with electric ones reduces the amount of CO2 being released into the atmosphere on a daily basis. The Achilles heel of electrical transportation lies in the car battery management system (BMS) that brings challenges to lithium-ion (Li-ion) battery optimization in finding the trade-off between driving and battery health in both the long- and short-term use. In order to optimize the state-of-health (SOH) of the EV battery, this study focuses on a review of the common Li-ion battery aging process and behavior detection methods. To implement the driving behavior approaches, a study of the public dataset produced by real-world EVs is also provided. This research clarifies the specific battery aging process and factors brought on by EVs. According to the battery aging factors, the unclear meaning of driving behavior is also clarified in an understandable manner. This work concludes by highlighting some challenges to be researched in the future to encourage the industry in this area.

/pdf/recognition-of-driving-behavior-in-electric-vehicles-li-ion-se1as3sj.pdf

Recognition of Driving Behavior in Electric Vehicle’s Li-Ion Battery Aging

Complex encounter situation modeling and prediction method for unmanned ships based on bounded rational game

Aggressive driving behavior (ADB) is one of the main causes of traffic accidents. The accurate recognition of ADB is the premise to timely and effectively conduct warning or intervention to the driver. There are some disadvantages, such as high miss rate and low accuracy, in the previous data-driven recognition methods of ADB, which are caused by the problems such as the improper processing of the dataset with imbalanced class distribution and one single classifier utilized. Aiming to deal with these disadvantages, an ensemble learning-based recognition method of ADB is proposed in this paper. First, the majority class in the dataset is grouped employing the self-organizing map (SOM) and then are combined with the minority class to construct multiple class balance datasets. Second, three deep learning methods, including convolutional neural networks (CNN), long short-term memory (LSTM), and gated recurrent unit (GRU), are employed to build the base classifiers for the class balance datasets. Finally, the ensemble classifiers are combined by the base classifiers according to 10 different rules, and then trained and verified using a multi-source naturalistic driving dataset acquired by the integrated experiment vehicle. The results suggest that in terms of the recognition of ADB, the ensemble learning method proposed in this research achieves better performance in accuracy, recall, and F1-score than the aforementioned typical deep learning methods. Among the ensemble classifiers, the one based on the LSTM and the Product Rule has the optimal performance, and the other one based on the LSTM and the Sum Rule has the suboptimal performance.

/pdf/a-recognition-method-of-aggressive-driving-behavior-based-on-35uzuiuu.pdf

A Recognition Method of Aggressive Driving Behavior Based on Ensemble Learning

Driving propensity is the driver’s attitude towards the actual traffic situation and the corresponding decision-making or behavior during the driving process. It is of great significance to improve the accuracy of safety early warning and reduce traffic accidents. In this paper, a real-time identification system of driving propensity based on AutoNavi navigation data is proposed. The main work includes: (1) A dynamic data acquisition method of AutoNavi navigation is proposed to obtain the time, speed and acceleration of the driver during the navigation process. (2) The dynamic data collection method of AutoNavi navigation is analyzed and verified through the dynamic data obtained in the real vehicle experiment. The principal component analysis method is used to process the experimental data to extract the driving propensity characteristics variables. (3) The fruit fly optimization algorithm combined with GRNN (generalized neural network) and the feature variable set are used to build a FOA-GRNN-based model. The results show that the overall accuracy of the model can reach 94.17%. (4) A driving propensity identification system is constructed. The system has been verified through real vehicle test experiments. This paper provides a novel and convenient method for building personalized intelligent driver assistance systems in practical applications.

/pdf/a-real-time-recognition-system-of-driving-propensity-based-kes0u4du.pdf

A Real-Time Recognition System of Driving Propensity Based on AutoNavi Navigation Data

Accurate identification of abnormal driving behavior is very important to improve driver safety. Aiming at the problem that threshold or traditional machine learning methods are mostly used in existing studies, it is difficult to accurately identify abnormal driving behavior of vehicles, a method of abnormal driving behavior recognition based on smartphone sensor data and convolutional neural network (CNN) combined with long and short-term memory (LSTM) was proposed. Smartphone sensors are used to collect vehicle driving data, and data sets of various driving behaviors are constructed by preprocessing the data. A recognition model based on a convolutional neural network combined with a long short-term memory network was constructed to extract depth features from data sets and recognize abnormal driving behaviors. The test results show that the accuracy of the model based on CNN-LSTM can reach 95.22%, and the performance indexes can reach more than 94%. Compared with the recognition model constructed only by CNN or LSTM, this model has higher recognition accuracy.

/pdf/abnormal-driving-behavior-recognition-method-based-on-smart-1895r0it.pdf

Abnormal Driving Behavior Recognition Method Based on Smart Phone Sensor and CNN-LSTM

: Accurate identification of abnormal driving behavior is very important to improve driver safety. Aiming at the problem that threshold or traditional machine learning methods are mostly used in existing studies, it is difficult to accurately identify abnormal driving behavior of vehicles, a method of abnormal driving behavior recognition based on smartphone sensor data and convolutional neural network (CNN) combined with long and short-term memory (LSTM) was proposed. Smartphone sensors are used to collect vehicle driving data, and data sets of various driving behaviors are constructed by preprocessing the data. A recognition model based on a convolutional neural network combined with a long short-term memory network was constructed to extract depth features from data sets and recognize abnormal driving behaviors. The test results show that the accuracy of the model based on CNN-LSTM can reach 95.22%, and the performance indexes can reach more than 94%. Compared with the recognition model constructed only by CNN or LSTM, this model has higher recognition accuracy. Abstract : A garment sizing system is essential for effective clothing design and production. A sizing system classifies a specific population into homogeneous subgroups based on some key dimensions. Persons of the same subgroup have the same body shape characteristics, and share the same garment size. Anthropometric data plays important role in creating clothing sizing system. The current work represents the sixth step towards the overall goal of developing the Libyan children’s clothing standards system based on physical measurements of the human body of Libyan schoolchildren. The objective of the current work is to study the physical measurements of students aged 6 to 17 years in the stages of primary, secondary. The body measurements of school children in Benghazi were collected and analyzed using simple statistics methods to understand the body ranges and current of student in all stages to develop the system sizing. The measurements were collected from previous projects. Some measurements were collected to complement a work of 90 (male and female) students between 6, 7 and 8 years old from a school in Benghazi. ANOVA test was used to determine differences between age groups. Abstract : Aiming at the problems of low detection accuracy and poor recognition effect of small-scale targets in traditional vehicle and pedestrian detection methods, a vehicle and pedestrian detection method based on improved YOLOv4-Tiny is proposed. On the basis of YOLOv4-Tiny, the 8-fold down sampling feature layer was added for feature fusion, the PANet structure was used to perform bidirectional fusion for the deep and shallow features from the output feature layer of backbone network, and the detection head for small targets was added. The results show that the mean average precision of the improved method has reached 85.93%, and the detection performance is similar to that of YOLOv4. Compared with the YOLOv4-Tiny, the mean average precision of the improved method is increased by 24.45%, and the detection speed reaches 67.83FPS, which means that the detection effect is significantly improved and can meet the real-time requirements. Abstract : The clayey soils in areas with seasonal frost are exposed to at least one freeze-thaw cycle every year and worsen their engineering properties. To prevent the engineering properties of clayey soils, it is necessary to improve the freeze-thaw resistance of them. In this study, the clayey soil was stabilized by using red mud and cement additive materials. Prepared samples of clayey soil and stabilized clayey soil were subjected to the unconfined compressive test. To investigate the effects of red mud and cement additive materials on the freeze-thaw resistance of clayey soil, the natural and stabilized expansive soil samples were exposed to the freeze-thaw cycles under laboratory conditions. The obtained results showed that the red mud and cement additive materials increased the freeze-thaw resistance of clayey soil. Consequently, it was concluded that red mud and cement additive materials can be successfully used to improve the freeze-thaw resistance of clayey soils. Abstract : The compacted clayey soils crack on drying because of their high swelling potential, and their hydraulic conductivities increase. To solve this problem, it is essential to stabilize the clayey soils using additive materials. The aim of this study is to examine the suitability of quartzite as a stabilization material to reduce the development of desiccation cracks in compacted clayey liner and cover systems. Experimental study was conducted to investigate the effect of wetting-drying cycles on the initiation and evolution of cracks in compacted clayey soils. For experimental studies, seven samples were prepared stabilized by using 0%, 2.5%, 5%, 7,5%, 10%, 12,5% and 15% quartzite and then they were subjected to four subsequent wetting-drying cycles. The results show that quartzite decreases the development of desiccation cracks on the surface of compacted samples. It is concluded that quartzite as a geological material can be successfully used to reduce the development of desiccation cracks in compacted clayey liner and cover systems exposed wetting-drying cycles. Abstract: Waste Management can be quite challenging, especially in the developing world. The challenge becomes even more complex with the growing population. City planners and decision makers are turning to technology to improve the efficiency of the waste management process. Geospatial technologies have offered a range of solutions, which have been deployed with success in waste management. This paper highlights the challenges of waste management in Port Harcourt, Nigeria, and how the Rivers State Waste Management Agency (RIWAMA) deployed the use of GIS in solving most of the challenges of identification, planning, evacuation, and transportation of wastes within the Port Harcourt metropolis. This paper looked at how this solution was deployed to solve key challenges as well as stimulate citizen participation in the waste management process. The paper concludes with a set of recommendations for expanding the potentials of its application.

Intelligent and connected driving (ICD) is an effective method of the vehicle’s active safety and traffic efficiency. Lane-changing is the critical part of ICD behavior and the basic content of traffic flow theory. Lane selection is the most important during lane-changing, which has a critical influence on traffic safety and efficiency. Focusing on the development of ICD technology, a lane selection model based on phase-field coupling and set pair logic is proposed. In this model, the method of phased-field coupling is used to comprehensively analyze the full information of lanes with the interaction between cluster vehicles considering driving propensity. The certain and uncertain human-vehicle-environment information is fully explored using the method of set pair logic. In the lane selection model, the lane utility and utility gains of lane-changing are calculated to select the candidate target lane. And the overall satisfaction is analyzed to determine the target lane as the final strategy. The proposed lane selection model constructed in this research fully considers the development of ICD technology and adapts to the requirement of an Intelligent and Connected Vehicle. This research could provide a theoretical basis for vehicle autonomous behavior decision-making and ICD. 

Hao Li

Papers

A Recognition Method of Aggressive Driving Behavior Based on Ensemble Learning

A Real-Time Recognition System of Driving Propensity Based on AutoNavi Navigation Data

Abnormal Driving Behavior Recognition Method Based on Smart Phone Sensor and CNN-LSTM

Abnormal Driving Behavior Recognition Method Based on Smart Phone Sensor and CNN-LSTM

Lane Selection Model Based on Phase-Field Coupling and Set Pair Logic