There different methods to measure canopy cover such as canopy analyzer, Lidar, remote sensing etc. The selection primarily depends on the area to be covered

The statistical of the differences between the techniques was tested with the non-parametric Kruskall-Wallis analysis of variance and multiple comparisons.

The results indicate that different techniques yield considerably different canopy

cover estimates. In general, labour intensive techniques provide unbased and more precise estimates, whereas the estimates provided by fast techniques (digital photographs, ocular estimation) have larger variances and may also be seriously biased. Plz check i have attached one publications also diffrent types.....useful to u..

Thank you all, I was actually going for a 20cmx20cm mirror with grid of 5x5cm. Wanted to know if someone has used it before. Densiometer was not available.

I have no idea regarding the method you mentioned;however we have used spherical densiometer and got satisfactory result in forest stands and tea plantations.We procured the equipment from the Roorkee based company-Virtual Electronics.

If we speak about statistical approach very promising could be point sampling. Good application of this method you can find on web-based tools ITree Canopy https://www.itreetools.org/canopy/index.php. 

Some ideas how to use remote sensing images of different spatial resolution you can find here: http://landcover.usgs.gov/pdf/canopy_density.pdf

There are one or two apps for cell phones that can attempt to estimate canopy cover. I tried one (canopy app) and there are others (percentage cover). You'll need to assess their usefulness yourself.

Canopy cover (CC) was obtained by the LAI with the following formula: CC = - 1. e 005  1 0 xp( ) - .6 LAI  1 2