Hi All
I am building a device that measures the diameter and change in diameter of a pupil when subject to a light stimulus.
The current setup is an enclosure which resembles a VR Headset, many IR LED's, an Infrared Camera, and an LED light source for stimulus.
I can successfully pick up the pupil, its change, and using an algorithm and image processing I can pick up the edge of the pupil during the video which outputs a changing ellipse.
The Issue I am having is related to the depth distance between the camera lens and the pupil of the individual. Because all individuals are different and have different facial structures, this depth will change by a few millimetres between each subject. This change can make two subjects, with the same pupil size, appear to have different pupil sizes (the subject who's pupil is further away from the camera, will appear to have a smaller pupil).
This slight change will introduce error into the accuracy of the pupil diameter calculation.
I haven't been able to find a paper that explains how the diameter is calculated when the depth is unknown?
If anyone can help, or point me in the right direction to solve this issue I am having, It will be greatly appreciated.
Kind Regards
Scott Bruton
Hi, did you solve your problem? A possible solution is to make your pc consider a normal corneal diameter is 12 mm, hence your device pc could compare pupil diameter to corneal diameter and calculate the former
- Badges
- Science topic
- Similar topics
- Computer Science
- Images