A Fuzzy-Logic Approach to Dynamic Bayesian Severity Level Classification of Driver Distraction Using Image Recognition (original) (raw)
Related papers
Computer Networks, 2018
This research paper presents a validated emotion enabled cognitive driver assistance model (EECDAM) as an accident prevention scheme while keeping in mind different types of driver distractions. It is observed that distracted drivers know that distraction can lead them to a crash but they are not aware of distractions when they take over and they continue to drive. With advancements in autonomous vehicles technologies, it is possible to have an onboard driver assistance systems. However, research is yet to be reported on this issue whether onboard driver assistance program will be effective or not. The Emotion Enabled Cognitive Driver Assistance Model is a system based on an encapsulated Emotion Enabled Cognitive Driver Assistant (EECDA), which computes the effects of external factors at the distraction level of the subject and generates algorithmically generated fear emotion. During experiments, the EECDA intervenes when the fear intensity of the driver crosses a threshold by sending two sound alerts to the driver to take appropriate action. To demonstrate the effectiveness of the proposed approach as a road safety system, a Cognitive Agent-Based Computing (CABC) framework has been utilized to validate the results of the EECDAM. Algorithms are utilized using fuzzy sets to compute distraction of the drivers. We also present an Agent-Based Model (ABM) to validate the implementation of the proposed scheme. Extensive experiments demonstrate the proficiency of the proposed model for robust collision avoidance.
Simultaneous analysis of driver behaviour and road condition for driver distraction detection
International Journal of Image and Data Fusion, 2011
The design of intelligent driver assistance systems is of increasing importance for the vehicle-producing industry and road-safety solutions. This article starts with a review of road-situation monitoring and driver's behaviour analysis. This article also discusses lane tracking using vision (or other) sensors, and the strength or weakness of different methods of driver behaviour analysis (e.g. iris or pupil status monitoring, and EEG spectrum analysis). This article focuses then on image analysis techniques and develops a multi-faceted approach in order to analyse driver's face and eye status via implementing a real-time AdaBoost cascade classifier with Haar-like features. The proposed method is tested in a research vehicle for driver distraction detection using a binocular camera. The developed algorithm is robust in detecting different types of driver distraction such as drowsiness, fatigue, drunk driving or the performance of secondary tasks.
Real-Time Detection System of Driver Distraction Using Machine Learning
IEEE Transactions on Intelligent Transportation Systems, 2013
There is an accumulating evidence that driver's distraction is a leading cause of vehicle crashes and incidents. In particular, it has become an important and growing safety concern with the increasing use of the so-called In-Vehicle Information Systems (IVIS) and Partially Autonomous Driving Assistance Systems (PADAS). Thereby, the detection of the driver status is of paramount importance, in order to adapt IVIS and PADAS accordingly, so avoiding or mitigating their possible negative effects. The purpose of this paper is to illustrate a method for the non-intrusive and real-time detection of visual distraction, based on vehicle dynamics data and without using the eye-tracker data as inputs to classifiers. Specifically, we present and compare different models, based on well-known Machine Learning methods. Data for training the models were collected using a static driving simulator, with real human subjects performing a specific secondary task (SURT) while driving. Different training methods, model characteristics and feature selection criteria have been compared. Based on our results, SVM has outperformed all the other ML methods, providing the highest classification rate for most of the subjects. Potential applications of this research include the design of adaptive IVIS and of "smarter" PADAS.
2016
In this work, a preliminary study for automatic detection of driver's distraction due to cognitive load has been done by analyzing driving data from driving simulator. The simulation experiments are done with 4 drivers and three type of driving situation (normal driving and driving with various types of cognitive loads). From the analysis of time series data obtained from sensors in a driving simulator, it has been noticed that driving behavior changes with statistical significance for varying cognitive tasks. The feature of the driving simulator data that changes most with increasing cognitive load has been assessed and the classification accuracy for driving with and without cognitive load by a simple classifier came out on the average as 66.3%
Look at the Driver, Look at the Road: No Distraction! No Accident!
2014 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2014
The paper proposes an advanced driver-assistance system that correlates the driver's head pose to road hazards by analyzing both simultaneously. In particular, we aim at the prevention of rear-end crashes due to driver fatigue or distraction. We contribute by three novel ideas: Asymmetric appearance-modeling, 2D to 3D pose estimation enhanced by the introduced Fermat-point transform, and adaptation of Global Haar (GHaar) classifiers for vehicle detection under challenging lighting conditions. The system defines the driver's direction of attention (in 6 degrees of freedom), yawning and head-nodding detection, as well as vehicle detection, and distance estimation. Having both road and driver's behaviour information, and implementing a fuzzy fusion system, we develop an integrated framework to cover all of the above subjects. We provide real-time performance analysis for real-world driving scenarios.
Detection of Driver Distraction Using Vision-Based Algorithms
The risk of drivers engaging in distracting activies is increasing as in-vehicle technology and carried-in devices become increasingly common and complicated. Consequently, distraction and inattention contribute to crash risk and are likely to have an increasing influence on driving safety. Analysis of police-reported crash data from 2008 showed that distractions contributed to an estimated 5,870 fatalities and 515,000 injuries. This paper assesses the extent to which vision-based algorithms can detect different types of driver distraction under different driving conditions. Data were collected on the National Advanced Driving Simulator from 32 volunteer drivers between the ages of 25 and 50. Participants drove through representative situations on three types of roadways (urban, freeway, and rural) twice: once with and once without distraction tasks. The order of the drives was counterbalanced. The three distraction tasks included a reaching task, a visual-manual task and a cognitive task which were repeated eight times throughout the drive.
Machine learning based classifier model for autonomous distracted driver detection and prevention
Recent researches and surveys have provided us with the evidence that distracted driver is a major cause of vehicle crashes all around the world. In-vehicle information systems (IVIS) have raised driver safety concern and thus, detecting distracted driver is of paramount importance. The project (or paper) shows a method of real-time distraction detection and initiates safety measures. In the realization of this project we have used Web-Cam, Raspberry Pi (a low cost, small size computing device), along with concepts of deep learning and convolutional neural networks. We classify drivers into multiple categories of distraction, some of them are texting, drinking, operating IVIS etc. Web-Cam feeds the classifier with real-time images of the driver of a particular vehicle. The system also constitutes a buzzer alarm which rings once the distraction is detected.
Video-based detection and analysis of driver distraction and inattention
2014 International Conference on Signal Processing and Integrated Networks (SPIN), 2014
This paper addresses two issues for mitigating driver distraction/inattention by using novel video analysis techniques: (a) inside an ego vehicle, driver inattention is monitored through first tracking drivers face/eye region using Riemannian manifoldbased particle filters, followed by recognition of dynamic eye states using PPCA (probabilistic principal component analysis) and SVM (support vector machine) classifier. Frequencies of eye blinking and eye closure are used as the indication of sleepy and warning sign is then generated for recommendation; (b) outside an ego vehicle, road traffic is also analyzed. Surrounding vehicles (in both directions) are tracked, and their states are analyzed by self-calibrated cameras using view-geometries and road information. Parameters (e.g. distance, velocity, number) of tracked vehicles are estimated on the road ground plane in the 3D world coordinate system. These pieces of information are provided for mitigating drivers inattention. The main novelties of the proposed scheme include facial geometry based eye region detection for eye closure identification, combined tracking and detection of vehicles, new formulae derived in camera self-calibration, and the hybrid system that handles both daytime and nighttime scenarios. Experiments have been conducted on video data in two different types of camera settings, i.e., captured inside and outside a vehicle. Preliminary tests have been conducted, results and performance evaluation have indicated the effectiveness of the proposed methods.
An approach to classify distraction driver detection system by using mining techniques
Indonesian Journal of Electrical Engineering and Computer Science, 2022
According to the motor vehicle safety division, over the past 5-10 years, usage of motor vehicles has rapidly increased, in that specifical usage of cars has grown tremendously. The major contribution of this paper is a systematic evaluation of the scholarly literature on driver distraction detection techniques. Our driver distraction detection framework offers a systematic overview of evaluated methodologies for detecting driver attention. So, we need to develop a model that classifies each driver's behaviour and determines its corresponding class name. To overcome this dispute, we have attained an appreciable number of deep learning algorithms on the dataset like convolutional neural network (CNN) and VGG16 to detect what the driver is doing in the car as given in the driver images. This process can be done by predicting the likelihood of the driver's actions in each picture. Of all models, we distinguished that the VGG16 Algorithm has conquered CNN with a loss of 0.298 and an Accuracy of 91.7%.