paper: Zebravision 4.0 Goal Detection - Team 900

retroreflective
zebravision
900
team900
zed
stereolabs
colormatching
zedcamera
goaldetection

#1

Thread created automatically to discuss a document in CD-Media.

Zebravision 4.0 Goal Detection - Team 900
by: ForeverAlon

Zebravision 4.0 is Team 900s vision system for the 2016 season; FIRST Stronghold. Our work was focused around recognizing the vision goals using shape and color based matching, recognizing the boulders using a neural network, and integrating the detection systems into a tracking system using the StereoLabs ZED stereo camera. One of the main features of Team 900’s Zebravision code this year was goal detection. This paper gives an overview of the hardware and code used.

Zebravision 4.0 is Team 900s vision system for the 2016 season; FIRST Stronghold. Our work was focused around recognizing the vision goals using shape and color based matching, recognizing the boulders using a neural network, and integrating the detection systems into a tracking system using the StereoLabs ZED stereo camera. One of the main features of Team 900’s Zebravision code this year was goal detection. This paper gives an overview of the hardware and code used.

The system used a Stereolabs ZED RGB-depth camera and green LED rings to highlight the retroreflective tape around the goal. The image was filtered to look for the reflected LED color and thresholded to turn it into binary “green / not green” image. The code then extracted contours from the image and applied a number of simple filters to rule out blobs which were obviously not goals. The remaining contours were scored in a number of criteria and the best scoring few objects were assumed to be goals. If more than one valid goal is found, several tiebreakers were used to pick one goal to shoot at. If a valid goal was found, the angle and distance to the target was reported; if none were found, a packet with -1.0 distance and angle was returned to the roboRIO.

Zebravision4.0GoalDetection.pdf (684 KB)


#2

Zebravision 4.0 is Team 900s vision system for the 2016 season; FIRST Stronghold. Our work was focused around recognizing the vision goals using shape and color based matching, recognizing the boulders using a neural network, and integrating the detection systems into a tracking system using the StereoLabs ZED stereo camera. One of the main features of Team 900’s Zebravision code this year was goal detection. This paper gives an overview of the hardware and code used.

The system used a Stereolabs ZED RGB-depth camera and green LED rings to highlight the retroreflective tape around the goal. The image was filtered to look for the reflected LED color and thresholded to turn it into binary “green / not green” image. The code then extracted contours from the image and applied a number of simple filters to rule out blobs which were obviously not goals. The remaining contours were scored in a number of criteria and the best scoring few objects were assumed to be goals. If more than one valid goal is found, several tiebreakers were used to pick one goal to shoot at. If a valid goal was found, the angle and distance to the target was reported; if none were found, a packet with -1.0 distance and angle was returned to the roboRIO.


#3

So I watched you guys practice and play at competition and I saw that you rarely missed if you decided to fire a boulder - your robot had lethal accuracy when there was no defender. I even saw small robots hit you and your turret would turn and you would score anyway.

My question is that with a defender in your way blocking your camera occasionally would you still be able to track the known location of the goal and fire? It seemed as though you could fire over most defenders even if they were 4ft 6in if you could track the location while they were blocking you and fire anyway. How did you combat camera obstruction?


#4

We didn’t. :frowning:

But we have plans for it now. We’re going to be adding latency and IMU compensation to the code to better track the targets. We will be working closely with Kauai Labs to make it happen before too long.