4a23a9274a
* builder: Build against Buster * builder: Use correct repository specifications * builder: Move ld.so.preload to have less errors * builder: Use coex repo to install Monkey * builder: Search for buster ROS packages * aruco_pose: Vendor in aruco library from OpenCV 3.4.6 * builder: Move to ROS Melodic * builder: Update kernel version * aruco_pose, clever: Remove opencv3 ROS dependency * builder: Update rosdep * travis: Disable eclint for vendored aruco library * tests: Don't try to locate opencv in ros * roscore: Use melodic distribution * Revert "aruco_pose: Vendor in aruco library from OpenCV 3.4.6" This reverts commit 9c14a8c002bb3396f9a7d9b2ba39969207f066ba. * aruco_pose: Vendor opencv_contrib/aruco again * builder: Add led packages * builder: Remove unused builder code * travis: Add native tests * builder: Set permissions for standalone-install * builder: Use -y for package installation * builder: Add repo for standalone build * builder: Use correct file types for standalone install * aruco_pose: Accept rgb8 map images * builder: Disable mjpg_streamer test * aruco_pose: Allow rgb8 map images (again) * builder: Re-add mjpgstreamer * builder: Install tornado==4.2.1 for rosbridge_suite * builder: Use more recent base image * builder: Use default kernel * builder: Move ld.so.preload back after tests * builder: Disable catkin tests These tests fail on a remote machine but seem to pass just fine on real hardware. Something must have changed between Kinetic and Melodic, and we must investigate more, but for now we just need a working image. * aruco_pose: Remove unused vendored code * selfcheck: Update systemd-analyze regex * builder: Add opencv repository * rosdep: Update package definitions for Melodic * rosdep: Use proper yaml formatting * travis: Remove unnecessary space * docs: Reference Melodic wherever possible
914 B
ArUco Marker Detection
ArUco
ArUco markers are easy to detect pattern grids that yield up to 1024 different patterns. They were built for augmented reality and later used for camera calibration. Since the grid uniquely orients the square, the detection algorithm can determing the pose of the grid.
ChArUco
ArUco markers were improved by interspersing them inside a checkerboard called ChArUco. Checkerboard corner intersectionsa provide more stable corners because the edge location bias on one square is countered by the opposite edge orientation in the connecting square. By interspersing ArUco markers inside the checkerboard, each checkerboard corner gets a label which enables it to be used in complex calibration or pose scenarios where you cannot see all the corners of the checkerboard.
The smallest ChArUco board is 5 checkers and 4 markers called a "Diamond Marker".