This project was started as a Science Fair project for Cross Roads Elementary School in Irmo, SC (https://www.lexrich5.org/cris) by Dan (http://dan.techdozor.org). Main focus of this project is obtaining an image, feeding it to the Vision API, and retrieving the identified properties of the image as a result of recognition operation. This project leverages node.js, Vision API from Google (possibly other providers), and even integrate with Alexa (see below).
- node.js
- export GOOGLE_APPLICATION_CREDENTIALS - pointing to json API Service Account https://cloud.google.com/vision/docs/auth. Note: have to go through motion of setting up the account, enabling the API, etc. etc. etc.
- don't forget to run npm install.
- First hit with your browser /upload URL (wherever you host this service)
- Upload your image (other files won't work)
- After call /results and see the results of the image recognition that today is based on Google Vision API and is able to recognize lables, text, labels.
Power your Alexa to see: https://github.com/techdozor/alexvisionskill
Have fun!