Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Occam Exposure Calibration #24

Open
belgiumkansas opened this issue Feb 8, 2019 · 2 comments
Open

Occam Exposure Calibration #24

belgiumkansas opened this issue Feb 8, 2019 · 2 comments
Assignees
Labels
Critical the upmost importance enhancement New feature or request

Comments

@belgiumkansas
Copy link
Contributor

Currently the Occam white balance and exposure settings with vary colors considerably depending on lighting conditions and whats in frame. We need a way to keep consistent colors for CV.

Current thinking is come kind of color calibration card on the nose of the sub where it is in frame of Occam camera 0. Then that gets feed into a node running on the sub that makes sure the Occam parameters are set in such a way that we have consistent color for CV as it white balances etc.

@belgiumkansas belgiumkansas transferred this issue from another repository Feb 15, 2019
@skhadem skhadem added the enhancement New feature or request label Mar 11, 2019
@skhadem
Copy link
Member

skhadem commented Aug 12, 2019

Lol this issue has been a pain for so long

@skhadem skhadem changed the title Occam color calibration. Occam Exposure Calibration Aug 12, 2019
@skhadem skhadem added the Critical the upmost importance label Aug 12, 2019
@flamma7 flamma7 self-assigned this Aug 28, 2019
@DrYerzinia
Copy link
Member

DrYerzinia commented Sep 20, 2019

Potential solutions:

  • Control loop that adjusts the gain to set the intensity of a know object like the nose of the sub to a predetermined value. This may have issues in conditions where the obstacle is directly illuminated but the sub is in darkness.
  • OCCAM has built in auto exposure but this tends to set it to be a bit bright, see if we can adjust it so it gives better results
  • Consider the mean value of the pixels of interest in the scene. This might be compute intensive. Do every few frames? All cameras use the same exposure so we cant do this on a per camera basis.
  • Use a histogram approach to change the light curve to minimize overexposure.

As part of this we should look at how we can adjust its target goal based on data collected from an autonomous run.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
Critical the upmost importance enhancement New feature or request
Projects
None yet
Development

No branches or pull requests

4 participants