raw images topic output with colour contrast

Hello,

I would like to know if there is a way to apply contrast to an image raw topic. Or if there is a package which inputs the raw images and outputs user-configured/edited raw images to another topic.

I have node which publishes coloured images, and then I use orbslam2 for SLAM mapping (which processes automatically the images into greyscale). The problem is that the greyscale image still has many information I don't want to map. So the solution I have thought of is to give more contrast to the images published by the camera node in order to reduce the details that orbslam2 sees when mapping.

Thank you for your help!

Asked by MrRivi on 2020-08-24 05:37:33 UTC

Comments

Answers

ROS itself does not provide image processing capabilities, what you could do is to use OpenCV to process the image, then republish it to another topic of your choice.

Have a look at the cv_bridgepackage to see how to convert ROS image formats to OpenCV image formats and back. Here are some tutorials

Asked by rfn123 on 2020-08-24 07:47:20 UTC

Comments

Understood. Looking at the tutorials you mentioned and searching a bit more on my own I found this package: opencv_apps . It provides various nodes that run internally OpenCV's functionalities and publish the result as ROS topics. I think it can be used a workaround for my issue. Also image_proc is useful. And finally if the camera node can be dynamically reconfigured rqt_reconfigure can be a solution.

Asked by MrRivi on 2020-08-25 10:55:59 UTC