First of all, definitions. By synaptic strength let's understand ability of post-synapse to elicit current flow/depolarization in response to neurotransmitter appearance in the synapse. How can NT appear? By release from pre-synapse or through uncaging (optical activation of "caged", or inactive NT).
Second, post-synapse contain all sort of stuff, but main sources of current/depolarization are ionotropic receptors of NTs, e.g. NMDA-R or AMPA-R. This molecules are not just floating ion membrane but attached to extensive scaffolding just below of the membrane. Scaffolding includes actin, and more importantly PSD95 (in case of excitatory synapses).
I work in vivo and don't do any EM. But in vivo issue of observing molecules is that you can't label sensitive proteins without disturbing function. Some things are just too delicate to be labeled with fusing GFP.
PSD95 has been used for in vivo labeling with GFP. One paper I will point you to, is this knock-in work in mice: Live Imaging of Endogenous PSD-95 Using ENABLED: A Conditional Strategy to Fluorescently Label Endogenous Proteins, J Neurosc. 2014
This is the figure that shows strong correlation between brightness of synapse (aka number of PSD95 molecules) and current that occurs in cell body when authors uncaged glutamate at different synapses.
Now, how does that relates to your image? PSD stands for "post-synaptic density" protein, because in EM synapse is very dark. I don't know links, but I would assume some correlation between brightness of synapse in EM and its strength.