Projection of EMsoft master patterns from stereographic to Lambert projection struggles with equator

[hakonanes]

The projection of an EMsoft master pattern in the stereographic projection to the Lambert projection gives incorrect intensities from vectors on the equator. The below image table shows Lambert projection generated with kikuchipy from an EMsoft master pattern in the stereographic projection in the first column, and a Lambert master pattern created directly in EMsoft (not going from the stereographic projection) in the second column. The second row shows the upper left corner from both master patterns.

kikuchipy	EMsoft

The blacke edge pixels produced with kikuchipy is a result of the EMsoft master pattern having 0-intensities outside the equator, which are included in most of the edge pixels in the bi-quadratic interpolation of vector intensities on the equator. I don't know to solve this.