Barrel Detection

Approaches

The following approaches have been considered:

Find orange pixels (on barrel) with (Pixels) => Red - Green => High-Pass Threshold => (Orange pixels)
Find while pixels (on barrel) with (Pixels) => Saturation => Low-Pass Threshold => (White pixels)
(Orange pixels) => Blob seperation => Orange stripes (Bounding Boxes)
(Orange pixels) => Blob seperation => White stripes (Bounding Boxes)
{Orange stripes (Bounding Boxes), White stripes (Bounding Boxes)} => Blob merging => Barrels (Bounding Boxes)

Find orange pixels (on barrel) with (Pixels) => Red - Green => High-Pass Threshold => (Orange pixels)
Scan horizontal lines of image, starting from the bottom, and moving up. When currect scan line intersects a sufficient number of orange pixels, mark the orange segment of the line as being the bottom edge of a barrel. Once the bottom edge of a barrel is identified, ignore all pixels above the bottom edge. Continue scanning.

I have written an experimental image filter called BarrelBlobFinder, included in ImageFilterDemo v1.2.3, that implements a more complex version of this algorithm. It works quite well, although it could benefit from some more calibration. --David 23:22, 19 Oct 2005 (EDT)

Scan distances in front of robot by some method, obtaining a distance-map.
Find/distinguish the "plateaus" of the graph. Mark each "plateau" as the near-edge/near-face of a barrel.

SICK
- Obtains a distance-map along a line projected away from the robot.
- Easy to implement programmatically.
- Expensive hardware. May be able to get from the CS department
Stereoscopic Imaging
- Obtains a distance-map along a plane projected away from the robot.
- Difficult to implement programmatically.
- Requires two camcorders. Currently, we only have one.

In the above steps, items surrounded with parentheses are data. Other items are analyses or filters.