Hi everyone! I have just developed a smartphone app that measures light flickering. It measures the difference in percent from the highest light output to the lowest.
I am very curious about what you think! If you are interested, please try it out and let me know if it is useful and if there are any improvements needed. Thanks!
I tried it on Android, I don't think it is that sensitive or polling right. I was able to determine hand movements causes it to think flicker is occurring the more exaggerated the movement, the higher the flicker, so really it needs to be placed on a stable surface and not held at all.
I question some of the numbers too.. it was reporting 2% flicker on a T12 CCFL that has an electronic ballast at 20 kHz. I don't know that this number is accurate for my phone camera.
I took it to a closet that has a single incandescent bulb which has a current flicker of 120 Hz but due to blackbody effects it is not on and off like LEDs are. On the top shelf closest to it, it was reporting 2% flicker. Moving it to a lower shelf where there was less light but still the same bulb, it reported 5%.
I don't think a phone camera is capable of being a measuring instrument for this sort of thing.
Thanks a lot Sunspark !
Yes absolutely, it must be standing still - otherwise it will report a too high value.
Yes, that it reports between 2% and 5% on the same bulb is clearly a measurement error. We cannot expect it to be absolutely exact, so a few percentage points it can differ.
I have some LED bulbs at home that stays within 10% flickering, while another brand have a whopping 90% flickering all the time independent of how I am measuring. They both look the same to my eyes - but the app reveals a large difference in flickering. Would it never be valuable to measure this difference?
How are you polling? Are you requesting raw_sensor through the camera2 api?
