Some questions about GRIP

[iSpotix]

Our team has been learning GRIP and I've gotten it to publish data to the Network Tables on localhost. I think we want to run it on our Driver Station.

1. How can I find the IP address of the Network Tables server created by the roboRIO?

2. Is there a way to test for rectangularity? Either through opencv or using the perimeter and area published by GRIP.

http://imgur.com/gallery/5DmVBAM/
http://imgur.com/gallery/5DmVBAM/

Say if I had this image and I wanted to filter out the other target or other extraneous blobs that we detect.

[ThomasClark]

Our team has been learning GRIP and I've gotten it to publish data to the Network Tables on localhost. I think we want to run it on our Driver Station.

1. How can I find the IP address of the Network Tables server created by the roboRIO?


You don't need to specify this - just set the team number and GRIP will figure it out. GRIP v1.1.0 will make this clearer by actually filling in the text in the settings dialog instead of just silently doing it.


2. Is there a way to test for rectangularity? Either through opencv or using the perimeter and area published by GRIP.

http://imgur.com/gallery/5DmVBAM/
http://imgur.com/gallery/5DmVBAM/

Say if I had this image and I wanted to filter out the other target or other extraneous blobs that we detect.

GRIP v1.1.0 (which will probably be released within the next 24 hours) has a new solidity filter (https://github.com/WPIRoboticsProjects/GRIP/pull/378) that should help with that.

[Turing'sEgo]

Check to see if GRIP has approxpoly(DP). OpenCV has it, but I am not certain about GRIP. As the name suggests, it approximates a polygon out of a contour given an epsilon tolerance. You can read about the algorithm here (https://en.wikipedia.org/wiki/Ramer%E2%80%93Douglas%E2%80%93Peucker_algorithm).