Camera-Based Driver Monitoring System for Abnormal Behavior Detection
DOI: 10.5455/jjee.204-1586348076
archive: archived pipeline: cataloged verified
Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)
Summary
This paper addresses the critical safety issue of driver distraction and abnormal behavior, which contributes significantly to global road accidents. Motivated by the need for effective Driver Assistant Systems (DAS) and Driving Safety Support Systems (DSSS), the authors propose a camera-based monitoring system that detects abnormal driving behavior by analyzing the sequential patterns of a driver’s peripheral visual scanning. The core assumption is that a driver’s psychological and physiological status directly impacts their visual attention, and deviations from normal gaze patterns indicate distraction or fatigue. The system operates through five sequential stages implemented on an NVIDIA Jetson Nano platform. First, an in-vehicle camera captures video of the driver. Second, head pose is estimated using a deep learning model for facial landmark detection and Perspective-n-Point algorithms, with data smoothed by a Kalman filter. Third, gaze estimation is performed by detecting the iris center via pixel intensity thresholds. Fourth, a linear Support Vector Machine (SVM) classifier processes feature descriptors of head pose and pupil location to categorize the driver’s attention into one of six specific gaze zones: road, right mirror, left mirror, rear mirror, dashboard, and center console. Finally, an Echo State Network (ESN), a type of reservoir computing recurrent neural network, classifies the sequence of these gaze zones as either normal or abnormal behavior. The ESN was trained on simulated data comprising 4,000 sequences (2,000 normal and 2,000 abnormal), with normal behavior defined by specific time constraints on road glances and mirror checks. The experimental results demonstrate high efficacy in both classification stages. The linear SVM achieved a 99.2% accuracy in identifying gaze zones, outperforming other classifiers such as KNN and Naive Bayes. The ESN model successfully classified driving behavior with an accuracy of 98.3% on training data and maintained high performance on two independent test datasets, achieving accuracies of 98.0% and 98.25%. The system also incorporates alert rules triggered by prolonged abnormal behavior or excessive time away from the road or mirrors. The study concludes that analyzing temporal gaze patterns is a robust method for detecting driver distraction, with the proposed SVM and ESN combination offering high accuracy and computational efficiency. The system is deemed suitable for deployment in real-world DAS and DSSS applications. Future work aims to integrate this detection logic into embedded Advanced Driver Assistance Systems and expand the analysis to include mental concentration metrics and road environment features.
Provenance
The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed.
| Stage | Outcome | Tool | Model | Prompt | Attempts | Completed |
|---|---|---|---|---|---|---|
| discover | success | Crossref | — | — | 1 | 2026-06-25 |
| archive | success | canonical_url | — | — | 1 | 2026-06-26 |
| extract | success | cached | — | — | 2 | 2026-06-26 |
| clean | success | clean | — | — | 1 | 2026-06-26 |
| chunk | success | chunk | — | — | 1 | 2026-06-26 |
| embed | success | embed | Qwen/Qwen3-Embedding-8B | — | 1 | 2026-06-26 |
| enrich | failed | — | — | — | 1 | 2026-06-26 |
| promote | success | — | — | — | 1 | 2026-06-25 |
| summarize | success | llm | qwen3.6-27b-prismaquant | summ-v5 | 1 | 2026-06-26 |
| tag | success | vector_similarity | — | — | 6 | 2026-06-26 |
| verify | partial | — | — | — | 1 | 2026-06-26 |
Summary generated by qwen3.6-27b-prismaquant on 2026-06-26; verification: verified_with_issues.
Topics
Ranked by relevance to this paper. Hover a topic for its definition.
- gaze based attention detection
- distraction detection algorithms
- drowsiness detection algorithms
- dms validation
- eye movements scanning
- attention allocation
Information type
What kind of knowledge this paper contributes, grouped by family — independent of topic (what it is about) and method (how it was studied).
- Methodological Resource: tool software, measurement protocol
- Theoretical Contribution: computational model