Optalert Drowsiness Monitoring
Current drowsiness monitoring systems regularly sound annoying false alarms. Consumers will never accept this. Automakers that offer accurate, reliable systems will win in the new regulatory environment.
Optalert’s premium drowsiness solution provides the best of both worlds: Enhanced safety with no annoying false alerts. It uses our proprietary Johns Drowsiness Scale (JDS) that directly measures impairment from drowsiness.
European technical services offer testing for GSR and Euro NCAP compliance using Optalert’s JDS
The most accurate algorithm that measures impairment due to drowsiness
Can be integrated into any driver monitoring system (DMS) in days
Only needs one input signal: a driver’s eyelid movements (blinks)
Minimise false alerts – customers enjoy premium safety with no unnecessary disturbances
Patented JDS measures cognitive impairment, unlike KSS which only measures subjective tiredness
Scientifically validated in laboratories for over 20 years
Commercially validated in industrial fleets for over 15 years
Quantify drowsiness in the blink of an eye
Optalert’s patented drowsiness detection algorithm needs one input signal: the blinks of the driver’s eyelid. It then processes 64 characteristics of this signal to output a value from 0.0 to 10.0 on the Johns Drowsiness Scale (JDS), which spans the full spectrum of drowsiness, from alertness to sleep.
Optalert integrates with any DMS
Prolong wakefulness with effective countermeasures
Other drowsiness monitoring systems miss dangerous late-stage drowsiness and only detect signs of sleep, such as microsleeps or head nods.
Optalert’s algorithm detects cognitive impairment due to drowsiness far in advance of these emergency situations. The vehicle can then enact countermeasures that reduce impairment and prolong wakefulness.
Contact us to learn more about Optalert’s algorithm and how it can enhance your driver monitoring system.
You can also put us to the test! Submit your drowsiness data set below to see how it compares to the gold standard.
Drowsiness SDK requirements
Feature | Minimum requirement |
|---|---|
Eyelid signal | Only need one eyelid signal. Using two eyes adds redundancy in case of occlusion of one eye. Accurate signal of eyelid opening: vertical displacement between upper and lower eyelids measured from eyelid apex. Opening can be relative with no absolute dimensions. Raw, unprocessed eyelid signal is optimal. “Cleaned” signals strip out crucial data that reveal information about a subject’s underlying neurological state. |
Frame rate | ≥ 30 fps |
Resolution | ≥ 1.5 MP for narrow field of view (varies for wide field of view) |
Measurement resolution | Typically requires at least 32 vertical pixels |
Camera placement | Best results directly in front of driver, but can work on cluster, dashboard, rear vision mirror, A-pillar, or elsewhere |
Invalid flag | If occlusions occur or the eyelid opening accuracy is compromised the signal should be marked invalid. |
In just three simple steps, validate your DMS with Optalert’s world leading drowsiness system.
1
Upload your first 5 files or 180 minutes of eye-opening data.
2
Receive assessment report from Optalert.
3
Upload more data and receive full verification report from Optalert.
Want to know more?
Enter your name, email address, and your message, and we’ll get back to you as soon as possible.
You can also upload your files straight away and we will provide you with a comprehensive assessment report.
Frequently Asked Questions
Data needs to be in CSV file format. You can upload 5 files or 180 minutes of eye-opening data in total for the complimentary initial assessment.
Please ensure that the file contains five distinct columns:
- Timestamp (in milliseconds)
- Eyelid Left (eyelid opening measure; relative floating-point value from 0-1 inclusive)
- Eyelid Left Valid (Boolean, 1 or 0)
- Eyelid Right (eyelid opening measure; relative floating-point value from 0-1 inclusive)
- Eyelid Right Valid (Boolean, 1 or 0).
Each row should equate to a video frame and the video frame rate should be 30 fps or higher.
The valid flag should be 1 when there is high confidence that the eyelid reading accurately represents the distance between the upper and lower eyelids.
The Report on the first 5 files or 180 minutes of data is complimentary. If you need further assessment beyond the free tier, the assessment is charged at USD$10 per minute of data assessed.
This depends on several factors including the quality and quantity of your submitted data. You should receive your report within 10 business days.