All studies are affected by the quality and quantity of data.
Since driving is significantly affected by the driver’s emotions, driver emotion recognition studies have been conducted for various purposes such as driving safety, adjusting vehicle dynamics, and emotion elicitation of drivers. This has increased interest in the development of driver emotion recognition systems. In recent decades, the use of data-driven state-of-the-art techniques such as deep learning has increased interest in and performance of human affect recognition. The proposed system is avaliable on GitHub. The proposed data collection system enables the construction of reliable large-scale datasets on real-world driving and facilitates research on driver emotion recognition. To demonstrate the validity of our collected dataset, we also provide case studies for statistical analysis, driver face detection, and personalized driver emotion recognition. Data collection was completed without any accidents for over 122 h of real-world driving using the system, which also considers the minimization of behavioral and cognitive disturbances. The proposed system includes a self-reportable HMI application into which a driver directly inputs their current emotion state. To overcome this limitation, we propose a data collection system that collects multimodal datasets during real-world driving. However, current driver emotion datasets are limited by inconsistencies in collected data and inferred emotional state annotations by others. As vehicles provide various services to drivers, research on driver emotion recognition has been expanding.
0 kommentar(er)
