You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
When I use the grey co-occurrence matrix with a vertical angle, I expected 90° (pi/2 radians) to refer to the pixel above the reference pixel, thus be a counterclockwise direction. But the code below implies it's a clockwise direction. The docs don't specify this, so is the clockwise direction correct?
That measures the offset from the central pixel. Since sin and cos both start out positive, at an angle of 0 we start with an offset right, and as we increase the angle we move down, ie clockwise.
imho, for 1.0, we should swap the sin and cos calls. This would align with row/col coordinates and we would end up with a counterclockwise rotation starting from the rows axis (axis 0). Thoughts @scikit-image/core?
@sdtaylor in the meantime, we would certainly welcome a clarification to the docstring!
Description
When I use the grey co-occurrence matrix with a vertical angle, I expected 90° (pi/2 radians) to refer to the pixel above the reference pixel, thus be a counterclockwise direction. But the code below implies it's a clockwise direction. The docs don't specify this, so is the clockwise direction correct?
Way to reproduce
Version information
The text was updated successfully, but these errors were encountered: