{{ message }}
Repository navigation
improve convexityDefects and externalContours algorithms #826
Description
Currently there are some issues with convexityDefects detection:
A mask like this:
[
[0, 0, 0, 1, 0, 0, 0],
[0, 0, 0, 1, 0, 0, 0],
[0, 0, 0, 1, 0, 0, 0],
[1, 1, 1, 1, 1, 1, 1],
[0, 0, 0, 1, 0, 0, 0],
[0, 0, 0, 1, 0, 0, 0],
[0, 0, 0, 1, 0, 0, 0],
]will give us these points:
[
{ column: 3, row: 0 },
{ column: 3, row: 1 },
{ column: 3, row: 2 },
{ column: 4, row: 3 },
{ column: 5, row: 3 },
{ column: 6, row: 3 },
{ column: 3, row: 4 },
{ column: 3, row: 5 },
{ column: 3, row: 6 },
{ column: 2, row: 3 },
{ column: 1, row: 3 },
{ column: 0, row: 3 }
]We get points that are unique but they don't exactly form a homogenous contour. This creates issues with finding convexity defects.
Another potential issue is that currently we use convex hull points that are extended for correct surface calculation. This is not really fitting for calculating defects.
Activity