I am not sure what you mean by quantify? There are several parameters you can quantify in a fluorescent image (intensity, no of cells, ratiometric measurement if more than one sets of fluorescent filters was used, portion of the image above SNR, or twice the SNR, or above pixel intensity standard deviation, etc.).
For example, you can compare fluorescence intensity in one region of interest (ROI) vs another (for example sane vs diseased tissues).
Make sure that your images are not saturated for quantification (to do this you can plot the no of pixels vs intensity histogram).
If you compare several tissue sections between temselves, make sure you acquire the data in the same conditions or that you correct for the change of conditions. If you modify images (subtract background, normalization, etc.) try to be consistent or taking this into account into your analysis. The ideal is to use a reference within the same tissue (ROI1/ROI2) or (Lambda1/lamda2) and than compare this "referenced measurment" between samples.
If you don't have an image processing/analysis software, you can use Fiji (a free, open source software) to visualize and analyse your images.
Ok, hope you will get the answer you are looking for, I am more familiar with 2-photons fluorescence images. Regardless of the image contrast mechanism same principles apply, you could perhaps try to look at filaments diameter, density, branching... for quantification.
I would agree with Suzie, that the same principles for the work with the image would apply, no matter if it's a fluorescence or a second harmonic generation. The physics behind the signal you get is different, but the ways to process the image are the same.
Maybe you can specify what exactly do you want to quantify?