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

update backprojection grass training image #58

Open
bsubei opened this issue Dec 12, 2014 · 3 comments
Open

update backprojection grass training image #58

bsubei opened this issue Dec 12, 2014 · 3 comments

Comments

@bsubei
Copy link
Contributor

bsubei commented Dec 12, 2014

make sure there are no white lines or orange barrels. This training image should only have grass (the old one has white lines).

This should be a 5-minute task

@bsubei bsubei self-assigned this Dec 12, 2014
@bsubei bsubei added this to the Navigate using line_detection milestone Dec 12, 2014
@bsubei
Copy link
Contributor Author

bsubei commented Dec 12, 2014

actually, turns out old training image doesn't have white lines. But it has orange barrels.

@bsubei
Copy link
Contributor Author

bsubei commented Dec 14, 2014

I decided to make a node that goes through rosbags and calculates the histogram of all the frames (maybe add a skip config param) and then saves it to a yaml. This way, the line detection nodes that need to do histogram backprojection can just read in this training histogram and just keep using it. (perhaps this node can then be expanded to train more complex machine learning algorithms, but we'll see)

@bsubei
Copy link
Contributor Author

bsubei commented Mar 20, 2015

histogram_calculator and backprojection nodes finished.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

No branches or pull requests

1 participant