You can do this using the built-in NI-SCOPE functions or the LabVIEW pulse waveform measurements VIs by tweaking the reference levels appropriately. However, given the noise level in your signal, you may not be happy with the result, since these measurements are done essentially on a point-to-point basis. A more robust option is to find the approximate location of the level crossing, take N points on either side of it, do a simple curve fit to this data, then use the curve fit to determine the actual level crossing. A line fit would work on the first part of your curve, an exponetial or parabolic fit on the second. A further optimization would be to use a robust fitting method, but it does not look like you need to go that far.
If you need more details on any of these methods, let us know.
