I think the discussion @jdmore linked is an excellent resource for this but long. To mention a few key points:
iNat can be used for uploading some research specimens, but isn’t intended for large scale digitization of scientific collections.
One issue with scientific collections being added to iNat is that many collections are already linked to GBIF, and, as iNat RG observations are added to GBIF as well, they may be “double-counted” in any analyses based on that data.
If adding observations from collections occurs on an occasional basis it’s probably ok, as long as the rights issues are addressed. Good examples might be particularly valuable observations, adding species not currently represented on iNat, etc.
If the collection you’re talking about is a personal collection not included in any institutional collections, I would guess it would be ok to make an account for that collection and post observations from specimens it to a limited extent, but it is kind of an “off-brand” use of iNat, as the focus of iNat is really documenting interactions between people and other organisms in nature.