Group Abstract Group Abstract

Message Boards Message Boards

10
|
21K Views
|
4 Replies
|
24 Total Likes
View groups...
Share
Share this post:

"SafeCracker" puzzle - a coffee table challenge

Posted 10 years ago
Attachments:
POSTED BY: Helen Shiells
4 Replies

Machine learning should definitely help here. Please see this example. It is also in the docs. You can get training data as:

digits = ExampleData[{"MachineLearning", "MNIST"}, "Data"];

or you also can build your own training set.

enter image description here

POSTED BY: Vitaliy Kaurov

At present the heaviest data in the result images are the rotation boundaries. As we know the rotation angle before the images are cut out, one could include a rotation dependent margin and crop the characters square. I expect that's the most significant possible improvement.

It's a mild cop out, but building a custom classifier from the images would definitely give the most reliable match.

POSTED BY: David Gathercole
Posted 10 years ago

That's brilliant, thanks David. I have also had this issue where TextRecognize won't recognise these individual numbers. I tried image rotation (with the individual pictures from your notebook) and it seems to think that "19" is "E":

output = TableForm[Rasterize[#, ImageSize -> {45}] & /@ # & /@ resultImagery]
recog = {}; AppendTo[recog, Table[TextRecognize[ImageRotate[output[[1, 1]][[1]], i]], {i, 6, 7, 0.05}]]

Maybe (as you suggested) a filter of some sort will help.

POSTED BY: Helen Shiells

This is a fun problem. I had a few nice ideas and these rapidly degraded into a notebook.

The concept is to pick out things looking like digits with morphological tools. One can then prune off items too far from the center, and fit the most extreme digits to a circle, constricting the circle to find the inner rings.

filter progression

fitting the first ellipse

The reality of the data isn't so nice. Not only are the rings at angles to the camera, but this angle changes with the vertical element, so different ellipses are required for each tier. I fear not accounting for this z axis drop within each ring is one of several weakness to the solution. I was quite pleased with it. Notebook attached.

output

Attachments:
POSTED BY: David Gathercole
Reply to this discussion
Community posts can be styled and formatted using the Markdown syntax.
Reply Preview
Attachments
Remove
or Discard