Real-time traffic sign detection and recognition using Raspberry Pi

Nowadays, the number of road accident in Malaysia is increasing expeditiously. One of the ways to reduce the number of road accident is through the development of the advanced driving assistance system (ADAS) by professional engineers. Several ADAS system has been proposed by taking into consideration the delay tolerance and the accuracy of the system itself. In this work, a traffic sign recognition system has been developed to increase the safety of the road users by installing the system inside the car for driver’s awareness. TensorFlow algorithm has been considered in this work for object recognition through machine learning due to its high accuracy. The algorithm is embedded in the Raspberry Pi 3 for processing and analysis to detect the traffic sign from the real-time video recording from Raspberry Pi camera NoIR. This work aims to study the accuracy, delay and reliability of the developed system using a Raspberry Pi 3 processor considering several scenarios related to the state of the environment and the condition of the traffic signs. A real-time testbed implementation has been conducted considering twenty different traffic signs and the results show that the system has more than 90% accuracy and is reliable with an acceptable delay.

Effectiveness Analysis on Human-Machine Information Interaction of Intelligent Highway

Different human-machine collaboration modes and driving simulation tests with the orthogonal method considered are designed for a series of typical intelligent highway landscapes. The feedback of drivers under different interaction modes is evaluated through NASA-LTX questionnaire, driving simulator, eye tracker, and electroencephalograph (EEG). This optimal interaction mode (including voice form, broadcasting timing, and frequency) of each driving assistance scene in CVI (Cooperative Vehicle Infrastructure) environment under the conditions of high and low traffic is determined from subjective and objective perspectives. In accordance with feedback of these subjects on each set scene, the voice information structure of each assistance mode plays the most important role on drivers followed by the broadcasting timing and frequency. These broadcasts which provide good effects include scenarios such as various assistance scenes at curves and an early warning timing at a long-distance trip as well as a high early warning frequency; in addition, as for an exit-tip assistance scenario, a voice mode assistance is preferred; and for various speed assistance scenes, the beep mode is better. Furthermore, it is found that, at a higher traffic level but a short-distance trip, an early warning timing is favored generally for various scenes while under a low traffic level, a long-distance early warning timing is better.

Real-Time Driving Behavior Identification Based on Multi-Source Data Fusion

Real-time driving behavior identification has a wide range of applications in monitoring driver states and predicting driving risks. In contrast to the traditional approaches that were mostly based on a single data source with poor identification capabilities, this paper innovatively integrates driver expression into driving behavior identification. First, 12-day online car-hailing driving data were collected in a non-intrusive manner. Then, with vehicle kinematic data and driver expression data as inputs, a stacked Long Short-Term Memory (S-LSTM) network was constructed to identify five kinds of driving behaviors, namely, lane keeping, acceleration, deceleration, turning, and lane changing. The Artificial Neural Network (ANN) and XGBoost algorithms were also employed as a comparison. Additionally, ten sliding time windows of different lengths were introduced to generate driving behavior identification samples. The results show that, using all sources of data yields better results than using the kinematic data only, with the average F1 value improved by 0.041, while the S-LSTM algorithm is better than the ANN and XGBoost algorithms. Furthermore, the optimal time window length is 3.5 s, with an average F1 of 0.877. This study provides an effective method for real-time driving behavior identification, and thereby supports the driving pattern analysis and Advanced Driving Assistance System.

Robust Pedestrian Detection Based on Multi-Spectral Image Fusion and Convolutional Neural Networks

Nowadays, pedestrian detection is widely used in fields such as driving assistance and video surveillance with the progression of technology. However, although the research of single-modal visible pedestrian detection has been very mature, it is still not enough to meet the demand of pedestrian detection at all times. Thus, a multi-spectral pedestrian detection method via image fusion and convolutional neural networks is proposed in this paper. The infrared intensity distribution and visible appearance features are retained with a total variation model based on local structure transfer, and pedestrian detection is realized with the multi-spectral fusion results and the target detection network YOLOv3. The detection performance of the proposed method is evaluated and compared with the detection methods based on the other four pixel-level fusion algorithms and two fusion network architectures. The results attest that our method has superior detection performance, which can detect pedestrian targets robustly even in the case of harsh illumination conditions and cluttered backgrounds.

Development of a Hybrid Method to Generate Gravito-Inertial Cues for Motion Platforms in Highly Immersive Environments

Motion platforms have been widely used in Virtual Reality (VR) systems for decades to simulate motion in virtual environments, and they have several applications in emerging fields such as driving assistance systems, vehicle automation and road risk management. Currently, the development of new VR immersive systems faces unique challenges to respond to the user’s requirements, such as introducing high-resolution 360° panoramic images and videos. With this type of visual information, it is much more complicated to apply the traditional methods of generating motion cues, since it is generally not possible to calculate the necessary corresponding motion properties that are needed to feed the motion cueing algorithms. For this reason, this paper aims to present a new method for generating non-real-time gravito-inertial cues with motion platforms using a system fed both with computer-generated—simulation-based—images and video imagery. It is a hybrid method where part of the gravito-inertial cues—those with acceleration information—are generated using a classical approach through the application of physical modeling in a VR scene utilizing washout filters, and part of the gravito-inertial cues—the ones coming from recorded images and video, without acceleration information—were generated ad hoc in a semi-manual way. The resulting motion cues generated were further modified according to the contributions of different experts based on a successive approximation—Wideband Delphi-inspired—method. The subjective evaluation of the proposed method showed that the motion signals refined with this method were significantly better than the original non-refined ones in terms of user perception. The final system, developed as part of an international road safety education campaign, could be useful for developing further VR-based applications for key fields such as driving assistance, vehicle automation and road crash prevention.

Modeling commercial vehicle drivers’ acceptance of advanced driving assistance system (ADAS)

Purpose Advanced driving assistance system (ADAS) has been applied in commercial vehicles. This paper aims to evaluate the influence factors of commercial vehicle drivers’ acceptance on ADAS and explore the characteristics of each key factors. Two most widely used functions, forward collision warning (FCW) and lane departure warning (LDW), were considered in this paper. Design/methodology/approach A random forests algorithm was applied to evaluate the influence factors of commercial drivers’ acceptance. ADAS data of 24 commercial vehicles were recorded from 1 November to 21 December 2018, in Jiangsu province. Respond or not was set as dependent variables, while six influence factors were considered. Findings The acceptance rate for FCW and LDW systems was 69.52% and 38.76%, respectively. The accuracy of random forests model for FCW and LDW systems is 0.816 and 0.820, respectively. For FCW system, vehicle speed, duration time and warning hour are three key factors. Drivers prefer to respond in a short duration during daytime and low vehicle speed. While for LDW system, duration time, vehicle speed and driver age are three key factors. Older drivers have higher respond probability under higher vehicle speed, and the respond time is longer than FCW system. Originality/value Few research studies have focused on the attitudes of commercial vehicle drivers, though commercial vehicle accidents were proved to be more severe than passenger vehicles. The results of this study can help researchers to better understand the behavior of commercial vehicle drivers and make corresponding recommendations for ADAS of commercial vehicles.

Drivers’ Intentions to Use Different Functionalities of Conditionally Automated Cars: A Survey Study of 18,631 Drivers from 17 Countries

A number of studies have investigated the acceptance of conditionally automated cars (CACs). However, in the future, CACs will comprise of several separate Automated Driving Functions (ADFs), which will allow the vehicle to operate in different Operational Design Domains (ODDs). Driving in different environments places differing demands on drivers. Yet, little research has focused on drivers’ intention to use different functions, and how this may vary by their age, gender, country of residence, and previous experience with Advanced Driving Assistance Systems (ADAS). Data from an online survey of 18,631 car drivers from 17 countries (8 European) was used in this study to investigate intention to use an ADF in one of four different ODDs: Motorways, Traffic Jams, Urban Roads, and Parking. Intention to use was high across all ADFs, but significantly higher for Parking than all others. Overall, intention to use was highest amongst respondents who were younger (<39), male, and had previous experience with ADAS. However, these trends varied widely across countries, and for the different ADFs. Respondents from countries with the lowest Gross Domestic Product (GDP) and highest road death rates had the highest intention to use all ADFs, while the opposite was found for countries with high GDP and low road death rates. These results suggest that development and deployment strategies for CACs may need to be tailored to different markets, to ensure uptake and safe use.