you can treat the honey samples by standard acetolysis followed by staining and the procedures are all describes in all standard books on pollen analysis for example
Faegri/Iversen(1989), Textbook of pollenanalysis, 4th ed. and
Moore/Webb/Collinson, Pollen Analysis (1991).
Most essential for determination is a pollen slide reference collection.
So no "harvest" is necessary, the pollen grains are normally well preserved in the honey and you will need only a small sample of some ml for acetolysis.
As Barth says, you should not use acetolysis, you run the risk of damaging some particles of interest and lose important information. Use it as a supplementary method for further identification of a specific grain.
Dear Dr Kamel, acetolysis is essential only if the samples have unwanted organic matters. It is important to put in consideration whether the pollen grains in the reference material is acetolysis or not, so as to help in reliable identification.
I am happy to know that many colleagues agree with me that the non-use of acetolysis is essential intending honey and pollen loads investigation. Acetolysis is usefull to recognize pollen grain morphology every time.