Validity of method

[inthesmall]

As this appears to be a picture taken of the original using a smartphone (camera flash visible, silhouette of photographer distinguishable), are the assumptions underpinning the analysis valid? Specifically, can the ratio of the semi-major to semi-minor axis be assumed to be equal to that of the original? Not suggesting that I don't think it is, I'm just curious. What sort of compound effect would a reasonable angular offset of the smartphone camera have?

[inthesmall]

Could any effect be compensated for using the labels on the image as a reference?