Hello to everyone.
I’m implementing a QBC HDR-compatible driver for the imx708 camera for Jetson Orin Nano.
The process is ongoing and I have some promising results but I’ve got some issues with color artifacts in HDR mode.
While diving into sensor theory, I came across some conclusions that the sensor itself did not set the correct value for the short-exposure and long-exposure step registers. Digital scale and sum on circuits level been wrong. So the artifacts may appear In the case of such an error. Raspberry solved this issue by some external module call. Is there anything similar from the Jetson side?
In any case, I would appreciate any advice or other tips or any data/manuals/pieces of code that can be useful.
Have reference to below topic for some configuration in ISP overrides file.
Thank you so much for the reply.
But this is a DOL-HDR mode that uses a digital overlap technique. I think that DOL-HDR and QBC-HDR are different things. In QBC HDR - we can set on a frame a different bunch of pixels with different exposures simultaneously. As a result, we will get HDR with different color intensities and contrasts on the same picture. DOL - uses a sequence of frames to combine different exposures and fuse them into one.
Please feel free to correct me if I am wrong.
Does this mode fuse short/long exposure already don’t need Jetson ISP to fuse them? If yes then we can’t help on it.
Thanks
No, it doesn’t unfortunately. Are there instruments or additional functionality to Quad Bayer re-mosaic broken line correction? And probably a silly question: Is it a good idea to try to apply DOL-HDR functionality to the sensor that has to work with QBC-HDR?
I don’t have idea for QBC-HDR can’t tell if able apply the DOL-HDR configure for it or not. BTW, current only support DOL-HDR/PWL-HDR tuning.
Thanks
This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.