jlengrand/coffee_voyance
1
0
Fork 0
mirror of https://github.com/jlengrand/coffee_voyance.git synced 2026-03-10 08:11:18 +00:00
Code Issues Packages Projects Releases Wiki Activity

Some more info

Browse Source
This commit is contained in:
julien Lengrand-Lambert
2015-10-03 11:56:31 +02:00
parent 8a25e3672b
commit a7fc94676a
1 changed files with 7 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

9
voyante.py
View File

@@ -49,9 +49,14 @@ for img_loc in img_locations:
# In this case, we imagine that template number 1 is the image taken by the user.
# We then match this template against our knowledge database (templates 1 to 3).
# Since we match an image against itself, the result should be obvious.
test_img = res_img[2]
# You can try changing the test_img, and you will see that we always find the template back
test_img = res_img[0]
templates = res_img
tmpl_match, max_res = find_match(test_img, templates)
print "The best template is : " + str(tmpl_match) + " with correlation value " + str(max_res)
show_binary_images(res_img)
show_binary_images(res_img)
# Current limitations :
# You can't match images against templates if the templates are bigger than the image
# You have to find a way to crop the image properly to avoid having the table too. This can be done by the user, or automated later.