Abstract:Objective. The scarcity of electroencephalogram (EEG) data, coupled with individual and scenario variations, leads to considerable challenges in real-world EEG-based driver fatigue detection. We propose a domain adaptation method that utilizes EEG data collected from a laboratory to supplement real-world EEG data and constructs a cross-scenario and cross-subject driver fatigue detection model for real-world scenarios. Approach. First, we collect EEG data from subjects participating in a driving experiment conducted in both laboratory and real-world scenarios. To address the issue of data scarcity, we build a real-world fatigued driving detection model by integrating the real-world data with the laboratory data. Then, we propose a method named cross-scenario and cross-subject domain adaptation (CS2DA), which aims to eliminate the domain shift problem caused by individual variances and scenario differences. Adversarial learning is adopted to extract the common features observed across different subjects within the same scenario. The multikernel maximum mean discrepancy method is applied to further minimize scenario differences. Additionally, we propose a conditional multikernel maximum mean discrepancy constraint to better utilize label information. Finally, we use seven rules to fuse the predicted labels. Main results. We evaluate the CS2DA method through extensive experiments conducted on the two EEG datasets created in this work: the SEED-VLA and the SEED-VRW datasets. Different domain adaptation methods are used to construct a real-world fatigued driving detection model using data from laboratory and real-world scenarios, as well as a combination of both. Our findings show that the proposed CS2DA method outperforms the existing traditional and adversarial learning-based domain adaptation approaches. We also find that combining data from both laboratory and real-world scenarios improves the performance of the model. Significance. This study contributes two EEG-based fatigue driving datasets and demonstrates that the proposed CS2DA method can effectively enhance the performance of a real-world fatigued driving detection model.

Structure Consistent Unsupervised Domain Adaptation for Driver Behavior Recognition

Towards Self-Similarity Consistency and Feature Discrimination for Unsupervised Domain Adaptation.

Attention-based Cross-Layer Domain Alignment for Unsupervised Domain Adaptation

A New Bidirectional Unsupervised Domain Adaptation Segmentation Framework

Domain Adaptive Driver Distraction Detection Based on Partial Feature Alignment and Confusion-Minimized Classification

Unsupervised Domain Adaptation Approach for Vision-Based Semantic Understanding of Bridge Inspection Scenes Without Manual Annotations

Unsupervised Domain Adaptation for Self-Driving from Past Traversal Features

Brain-Inspired Domain-Incremental Adaptive Detection for Autonomous Driving

Multi-level Domain Adaptation for Lane Detection

A Structure-aware Framework of Unsupervised Cross-Modality Domain Adaptation via Frequency and Spatial Knowledge Distillation

A Fine-Grained Unsupervised Domain Adaptation Framework for Semantic Segmentation of Remote Sensing Images

A cross-scenario and cross-subject domain adaptation method for driving fatigue detection

Survey on Unsupervised Domain Adaptation for Semantic Segmentation for Visual Perception in Automated Driving

Dynamic Distribution Alignment With Dual-Subspace Mapping For Cross-Subject Driver Mental State Detection

Learning Driver-Irrelevant Features for Generalizable Driver Behavior Recognition

Unsupervised Domain Adaptation for Remote-Sensing Vehicle Detection Using Domain-Specific Channel Recalibration

Pseudo Label Refinery for Unsupervised Domain Adaptation on Cross-dataset 3D Object Detection

Threshold-adaptive Unsupervised Focal Loss for Domain Adaptation of Semantic Segmentation

DSDFormer: An Innovative Transformer-Mamba Framework for Robust High-Precision Driver Distraction Identification

Multisource Adaption for Driver Attention Prediction in Arbitrary Driving Scenes

Unsupervised Domain Adaptation for Remote Sensing Image Semantic Segmentation Using Region and Category Adaptive Domain Discriminator