A Deep Learning-Based System for Driver Fatigue Detection
Bibliographic record
Abstract
Driver fatigue is still a principal cause of traffic accidents.While many ways allowing fatigue detection, a diversity of obstacles such as head position, luminosity, and facial expressions make it a very challenging problem.In this paper, we propose a hybrid approach using deep learning techniques to detect driver drowsiness by combining between structural and global classification methods.The structural method tracks eyes, eyebrows, and mouth movements to assess blink and yawning, for this purpose we calculate eye-opening and mouth-opening ratios relative to their width.Five parameters are extracted LEM (left eye movement), REM (right eye movement), LEB M (left eyebrow movement), REBM (right eyebrow movement), and MM (mouth movement), whereas the global method is based on Convolutional Neural Network (CNN) to describe the whole face.Eight-layer pre-trained Alexnet network is used to extract features and make classification of each frame.To do video classification, the five structural parameters, along with the global classification decision, are combined into a single vector to be input into Long-Short-Term Memory (LSTM) networks that is an improved version of Recurrent Networks.LSTM decision score is determined after running 150 steps, providing information about driver state Extensive Experiments are performed on a Driver Drowsiness Detection Dataset that contains subjects of different ethnicities.The experimental results show that the proposed method with the combined features improves drowsiness detection significantly as well as outperforms the state-of-the-art models in terms of drowsiness scores.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.005 | 0.002 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".