I would like a recommendations on how to read (take a photo and output a number) with all sorts of cheap digital displays, but as a baby-step, I need to be able to read the number that's displayed on LCD, for example, an 8-segment LCD display, not back-lit on a classic bike computer device etc.
Workflow:
Someone takes a photo of a device with an LCD display in it. As close to it as possible so that the pix contain mostly the display.
We would then push that image (Jpeg) to a Bluemix service, the service returns the number in text as it has seen (or tell us if it has not seen or gives a low confidence quality indicator).
The Question
Are there already Watson services that I can begin using, or at least begin exploring and implement some more processes on top?
Which Watson technology I could use. Can the character recognition do this?
Manabu Tokunaga WinguMD, Inc.
Answer by paulread (711) | Jan 28, 2017 at 03:23 AM
Hi @imanabu,
You can use the Visual Recognition API to do what you require. There is a Node-RED sample available in Watson Developer Cloud GitHub. https://github.com/watson-developer-cloud/node-red-labs/tree/master/starter-kits/accessibility Please check it out!
Is there a correlation between Similarity Search and classifier in Watson Visual Recognition? 3 Answers
Watson doesn't recognize text from 7-segment-displays 2 Answers
Can't get a Visual Recognition Classifier_ID to persist 2 Answers
Get version information of Visual Recognition API 2 Answers
Acceso a documentos en Google para someter a Watson Visual Recogniition 0 Answers