• Journal of the Institute of Convergence Signal Processing
• /
• v.13 no.3
• /
• pp.136-141
• /
• 2012

One of the main reasons for serious road accidents is driving while drowsy. For this reason, drowsiness detection and warning system for drivers has recently become a very important issue. Monitoring physiological signals provides the possibility of detecting features of drowsiness and fatigue of drivers. One of the effective signals is to measure electroencephalogram (EEG) signals and electrooculogram (EOG) signals. The aim of this study is to extract drowsiness-related features from a set of EEG signals and to classify the features into three states: alertness, drowsiness, sleepiness. This paper proposes a neural-network-based drowsiness detection system using Linear Predictive Coding (LPC) coefficients as feature vectors and Multi-Layer Perceptron (MLP) as a classifier. Samples of EEG data from each predefined state were used to train the MLP program by using the proposed feature extraction algorithms. The trained MLP program was tested on unclassified EEG data and subsequently reviewed according to manual classification. The classification rate of the proposed system is over 96.5% for only very small number of samples (250ms, 64 samples). Therefore, it can be applied to real driving incident situation that can occur for a split second.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.22 no.6
• /
• pp.768-773
• /
• 2012

One of the main reasons for serious road accidents is driving while drowsy. For this reason, drowsiness detection and warning system for drivers has recently become a very important issue. Monitoring physiological signals provides the possibility of detecting features of drowsiness and fatigue of drivers. One of the effective signals is to measure electroencephalogram (EEG) signals and electrooculogram (EOG) signals. The aim of this study is to extract drowsiness-related features from a set of EEG signals and to classify the features into three states: alertness, drowsiness, sleepiness. This paper proposes a drowsiness detection system using Linear Predictive Coding (LPC) coefficients and Support Vector Machine (SVM). Samples of EEG data from each predefined state were used to train the SVM program by using the proposed feature extraction algorithms. The trained SVM program was tested on unclassified EEG data and subsequently reviewed according to manual classification. The classification rate of the proposed system is over 96.5% for only very small number of samples (250ms, 64 samples). Therefore, it can be applied to real driving incident situation that can occur for a split second.

• Journal of Auto-vehicle Safety Association
• /
• v.8 no.1
• /
• pp.13-18
• /
• 2016

More than 80 percent of traffic accidents related with lane departure believed to be the result of crossing the lane due to either negligence or drowsiness of the driver. Lane-departure related accident in the highway usually involve high fatality. Even though LDWS is believed to prevent accident 25% and reduce fatalities by 15% respectively, its effectiveness in performance is yet to be confirmed in many aspects. In this study, the vehicle lateral locations relative to warning zone envelop (earliest and latest warning zone) defined in ISO standard, ECE and NHTSA regulations are compared with respect to various factors including delays, vehicle speed and vehicle heading angle with respect to the lane. Since LDWS is designed to be activated at the speed over 60 km/h, vehicle speed range for the study is set to be from 60 to 100 km/h. The vehicle heading angle (yaw angle) is set to be up to 5 degree away from the lane (abrupt lane change) considering standard for lane change test using double lane-change test specification. The TLC is calculated using factors like vehicle speed, yaw angle and reaction time. In addition, the effect of vehicle type and reaction time have been considered to assess LDWS safety.

• 제어로봇시스템학회:학술대회논문집
• /
• 1997.10a
• /
• pp.1856-1859
• /
• 1997

This paper describes the arousal measurement and the control system using fuzzy logic to prevent drowsy driving. Sugeno's method was used for fuzzy inference in this study. Membership function and rule base were determined form the modfied arousal level criteria. The output of fuzzy inference tracked well the change of subject's arousal level. When IRI(Inter-SIR interval) was under the 60sec, maximum output of three step warning method was medium sound, but that of fuzzy logic system was changed from medium to big. Furthermore, the output of the fuzzy inference was highly correlated with $N_{z}$(r=0.99). Therefore, the fuzzy inference method for evaluation and the control of arousal will be more effective at real driving sityation than three step warning method.ning method.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2021.07a
• /
• pp.519-520
• /
• 2021

본 논문에서는 딥러닝 기반의 얼굴인식과 Harr Cascade 분류기를 이용한 눈인식, 스마트워치를 매개로 한 심박수 측정을 활용하여 운전자 졸음운전 경고 시스템을 제안하였다. 제안하는 시스템은 PERCLOS 방법을 적용하여 운전자의 눈 감은 시간을 누적시켜 졸음 상태 유무를 판단하고, 스마트워치의 HR센서를 활용한 운전자의 심박수 값 모니터링을 진행하여 졸음 발생 시 경고음을 발생시켜 졸음운전으로 인한 교통사고를 예방할 수 있다.

• Transactions of the Korean Society of Automotive Engineers
• /
• v.21 no.5
• /
• pp.54-58
• /
• 2013

In recent days, vehicles have become equipped with electric systems that assist and help drivers driving safe by reducing possible accidents. LDWS(Lane Departure Warning System) and LKAS(Lane Keeping Assistant System) are involved in assist systems, especially for lateral motion of vehicles. Sudden and inattentive lateral motion of vehicles due to drivers' fatigue, illness, inattention, and drowsiness are major causes of accidents in highway. LDWS and LKAS provide drivers with warnings or assisting power to reduce any possibilities of accidents. In order to prevent or minimize the possibilities of accidents, lateral motion control of vehicles has been introduced in this research. DGPS/RTK(Differential Global Positioning System/Real Time Kinematics) and GIS(Geographic Information System) have been used to obtain the current position of vehicles and decide when activate controlling lateral motion of vehicles. The presented lateral motion control has been validated with actual vehicle tests.

• Journal of Biomedical Engineering Research
• /
• v.18 no.4
• /
• pp.491-498
• /
• 1997

This paper describes the arousal measurement and control system using fuzzy logic to prevent drowsy driving. Sugeno's method was used for fuzzy inference in this study. Arousal evaluation and control criteria were modified from result of Nz-IRI analysis depending on arousal sate. Membership function and rule base of fuzzy inference were determined from the modified arousal level criteria When lRl (Inter-SIR Interval) was shorter than 60sec, outputs of both methods were changed from small to big, but output of three step warning method was same level until the next warning range. Since output of fuzzy inference tracked well the change of subject's arousal level, problems of three step warning method could be overcome by fuzzy inference method Furthermore, the output of the fuzzy inference was highly correlated with Nz(r = 0.99). Therefore, the fuzzy inference method for evaluation and the control of arousal will be more effective at real driving situation than three step warning method.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2005.11a
• /
• pp.249-252
• /
• 2005

본 논문에서는 자동차를 운전 시에 운전자의 얼굴과 눈의 영역을 자동으로 검출하고 눈동자를 추적하여 운전자의 졸음 여부를 판단하는 효과적인 시스템 구현방법을 제안한다. 복잡한 배경에서 얼굴과 눈을 검출하는 방법은 Haar-like feature의 원리를 이용하고 졸음운전으로 판단하는 방법은 눈동자 영역의 특성과 눈동자의 검출 유무, 움직임 등의 인식을 통하여 졸음운전 경보시스템의 실용화에 대한 가능성을 확인한다.

• Korean Journal of Remote Sensing
• /
• v.21 no.4
• /
• pp.317-327
• /
• 2005

The flood is the most general and frequently occurs among natural disasters. Generally flood by the rainfall which extends superexcellently for the occurrence but flash flood from severe rain storm gets up an absurd drowsiness at grade hour. This paper aims to 1 hour real-time flash flood and predict possibility at the area where is the possible flood will occur from the rainfall hour mountain after acquiring data in GIS(Geographic Information System) base by GcIUH(Geomorphoclimatic Instantaneous Unit Hydrograph). The flash flood occurrence is set up at 0.5m, 0.7m and 1.0m in standard depth. And this study suggests standard flood alarm which designed by probable flood according to duration time. The research result shows real-time flash flood evaluation has the suitable standard in the basin when comparing with the existing official warning announcement system considering topographical information.