Driver drowsiness causes fatal driving accidents. Thermal imaging is a suitable drowsiness detection method as it is non-invasive and robust against changes in the ambient light. In this paper, driver drowsiness is detected by measuring the forehead temperature at the region covering the supratrochlear artery and also the cheek temperature. About 30 subjects drove on a highway in a driving simulator in two sessions. A thermal camera was used to monitor the facial temperature pattern. The subjects’ drowsiness levels were estimated by three human observers. The forehead and the cheek regions were located and tracked in each frame. The forehead and the cheek skin temperatures were obtained at three levels of drowsiness. The Support Vector Machine, the K-Nearest Neighbor, and the regression tree classifiers were used. From wakefulness to extreme drowsiness, the forehead skin temperature and the absolute cheek-forehead skin temperature gradient decreased by 0.46°C and 0.81°C, respectively. But the cheek skin temperature increased by 0.35°C in two sessions. The gradient difference is on average 50% higher than the forehead or the cheek temperature change alone. The results indicate that drowsiness can be detected with an accuracy of 82%, sensitivity of 85%, specificity of 90%, and precision of 84%. Driver drowsiness can be detected by monitoring changes in the forehead and the cheek temperature signal. Also, the temperature gradient can be used as a more robust and sensitive indicator of drowsiness.
The high number of fatal crashes caused by driver drowsiness highlights the need for developing reliable drowsiness detection methods. An ideal driver drowsiness detection system should estimate multiple levels of drowsiness accurately without intervening in the driving task. This paper proposes a multi-level drowsiness detection system by a deep neural network-based classification system using a combination of electrocardiogram and respiration signals. The proposed method is based on a combination of convolutional neural networks (CNNs) and long short-term memory (LSTM) networks for classifying drowsiness by concurrently using heart rate variability (HRV), power spectral density of HRV, and respiration rate signal as inputs. Two models, a CNN-based model and a hybrid CNN-LSTM-based model were used for multi-level classifications. The performance of the proposed method was evaluated on experimental data collected from 30 subjects in a simulated driving environment. The performance and the results of both models are presented and compared. The best performance for both three-level and five-level drowsiness classifications was achieved by the CNN-LSTM model. The results indicate that the three-level and five-level classifications of drowsiness can be achieved with 91 and 67% accuracy, respectively.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.