A real-time driver fatigue identification method based on GA-GRNN

Abstract

It is of great practical and theoretical significance to identify driver fatigue state in real time and accurately and provide active safety warning in time. In this paper, a non-invasive and low-cost method of fatigue driving state identification based on genetic algorithm optimization of generalized regression neural network model is proposed. The specific work is as follows: (1) design simulated driving experiment and real driving experiment, determine the fatigue state of drivers according to the binary Karolinska Sleepiness Scale (KSS), and establish the fatigue driving sample database. (2) Improved Multi-Task Cascaded Convolutional Networks (MTCNN) and applied to face detection. Dlib library was used to extract the coordinate values of face feature points, collect the characteristic parameters of driver's eyes and mouth, and calculate the Euler Angle parameters of head posture. A fatigue identification model was constructed by using multiple characteristic parameters. (3) Genetic Algorithm (GA) was used to find the optimal smooth factor of Generalized Regression Neural Network (GRNN) and construct GA-GRNN fatigue driving identification model. Compared with K-Nearest Neighbor (KNN), Random Forest (RF), and GRNN fatigue driving identification algorithms. GA-GRNN has the best generalization ability and high stability, with an accuracy of 93.3%. This study provides theoretical and technical support for the application of driver fatigue identification.

A Recognition Method for Road Hypnosis Based on Physiological Characteristics

“…Therefore, it is demonstrated that it is feasible to use these parameters to identify these driving behaviors, and relevant methods are provided as technical support. For example, we have previously tried using cameras to identify driver fatigue [ 31 ] and distraction [ 32 ], which is the work conducted on the basis of our research and understanding of driving fatigue and driving distraction. This study is an exploratory study.…”

Machine learning-based cognitive load prediction model for AR-HUD to improve OSH of professional drivers

“…The crossover rate of GA reached 0.8 (80%). Roulette wheel selection method was adopted to select individuals from the population for reproduction ( 12 ).…”

“…AR-HUD technologies have aroused significant attention in the HCI field over the past few years. AR refers to a technology that superimposes computer-generated virtual objects onto the real world, creating a mixed reality environment where the virtual and physical worlds coexist HUD ( 12 ). Besides, it is a display technology that projects information onto a transparent screen or a windshield, such that users are enbaled to view information without looking away from the road or the task at hand ( 13 ).…”

“…A novel cognitive load test was introduced, and a cognitive load database was generated. Variations were validated using statistical tests, and novel classification and regression tasks were introduced for machine learning ( 12 ). Becerra-Sánchez et al ( 24 ) uses n-back task as an auxiliary task to induce the cognitive load of the main task (i.e., driving) in three different driving simulation scenarios.…”

“…Consequently, OSH risks may be posed since the information presented on the display may not be easily customizable, making it difficult for professional drivers to receive the specialized information they require (7)(8)(9). Furthermore, the display is likely to increase cognitive load, particularly for drivers not accustomed to using AR-HUD technology, such that fatigue and other OSH issues can be triggered (10)(11)(12). Accordingly, designing AR-HUD interfaces that are tailored to the specific needs of professional drivers takes on critical significance in improving OSH.…”