AprilTag 16h5
AprilTag 16h5 holds 30 markers of
4×4 bits. In OpenCV it is
cv2.aruco.DICT_APRILTAG_16h5.
Generate a AprilTag 16h5 board
The numbers
| OpenCV name | cv2.aruco.DICT_APRILTAG_16h5 |
| Bits in a marker | 4 × 4 = 16 |
| Markers available | 30, with IDs 0 to 29 |
| Printed width | 6 cells, border included |
| Bits the decoder repairs | 2 |
| Cell size on a 100 mm marker | 16.7 mm |
| Pixels wanted in the image | 72 (floor 24) |
| Range for a 100 mm marker | about 2.3 m |
Range assumes a camera 1920 px wide, 60 degree horizontal field of view, which is a focal length of about 1663 pixels, and a marker seen straight on.
When to use it
Four bit markers have the widest cells at any printed size, so they read from the greatest distance and survive the most blur. They carry the least redundancy, so they are the easiest to misread. This is an AprilTag family. Pick it when the rest of the stack already uses AprilTag, for example in ROS. OpenCV reads it through cv2.aruco just like an ArUco dictionary. With 30 markers this suits a single board or a small set of tagged objects.
The rest of the family
| Dictionary | Markers | Bits repaired |
|---|---|---|
| AprilTag 16h5 | 30 | 2 |