Skip to content

babajob/richdocs

Repository files navigation

Welcome to Babajob RichDocs

This is a stand alone library that accepts photos (PAN Cards, Aadhaar Card, Driver's Licenses, etc) and audio files and parses them with Google and Microsoft's vision, OCR and speech reco APIs to determine the information therein and then calls various authentication APIs to verify PAN and/or Aadhaar numbers are real and match the Indian government's records. This exposes itself as a CRUD based webservice.

Setup steps

  1. Install Node 6.4
  2. Sync this Repo
  3. Nav to your repo folder
  4. Run npm install

Get API Creds for Firebase, Google Cloud APIs and Microsoft

Get the Creds .json files from Sean or create your own from google API cloud and Firebase

  1. Add to root directory: FirebaseServiceAccount.json

Google API Creds

be sure to change the name of your project below from babajobrichdocs

  1. Add to root directory: RichDocsCreds.json

TODO: Get Microsoft Creds

We currently call the Microsoft Vision API to get a well-framed thumbnail of whatever

TODO Optional: Get BetterPlace Authentication Creds

We use betterplace to verify Indian PAN and Aadhaar information

TODO Optional: Get CloudConvert Authentication Creds

We use cloudconvert to convert facebook audio clips to formats that the Google Speech Reco accepts

#To Start the Service

  1. Open git bash
cd c:
cd c:/Users/sean_/Node/richDocs
export GOOGLE_APPLICATION_CREDENTIALS="./RichDocsCreds.json"
export GCLOUD_PROJECT="babarichdocs"
nodemon

#Sample Commands To ensure the service is working: http://localhost:3001/api/richdocs

Show form to upload docs: http://localhost:3001/index.html

http://localhost:3001/richdocs/public/index.html
http://localhost:3001/richdocs/api/richdocs

To DELETE

curl -XDELETE http://localhost:3001/api/richdocs/-KRb9H7hJbIqRc54Egwv

About

Repo for our OCR and AI work

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published