treescore
Score a Christmas Tree based on uniformity of lights, shape of tree, and colors used.
Writeup with pretty pictures can be see at http://christianreimer.github.io/treescore/
Simple Example
>>> import treescore
>>> fname = 'path/to/image/of/tree.png'
>>> picker = treescore.RegressionColorPicker.from_file('model.data')
>>> scores, images, composite = treescore.judge.score(fname, picker, images=True)
>>> print(scores)
Scores(overall=65, led=85, shape=97, color=12, area=0.290468)
>>> treescore.judge.utils.display_img(composite)
>>>This should display an image such as the following (depending on your tree of
course)
Installation
OpenCV (http://opencv.org/) is used and you will need to install it. This will probably be the biggest challenge you will encounter. You can follow the instructions for python3 and OpenCV version 3 over at http://www.pyimagesearch.com/opencv-tutorials-resources-guides
Once you have OpenCV and the python bindings installed, the rest should be as easy as
$ mkvirtualenv treescore
$ git clone https://github.com/christianreimer/treescore.git
$ cd treescore
$ pip install -r requirements.txt
$ ./judge --help
$Todo
There is still a lot of work to do to accurately detect colors and hence mask out the tree itself.