Rekog-AI - Image Recognition

What is this?

A web app that allows you to use Amazon's rekognition AI service to analyse images.

The app is written in Node.js and uses the Amazon Rekognition AI service to perform the image recognition and analysis.

See https://aws.amazon.com/rekognition/ for details

Functionality

• upload image
• get textual description (labels) for image
• extract text and words from image
• get face details including sentiment analysis (happy, sad, calm, etc.)

Screenshots

Repo Contents

• /config = example aws creds config file
• /controllers = UI http controller and rekognition API functionality
• /images = sample images (put your own images into this directory)
• /models = ui data model
• /public = stylesheet, favorite icon, ui assets
• /screenshots - screenshot images for this readme
• /views = UI ejs components, assets and partials

Todo

• add video recognition functionality
• add functionality to capture screenshots from webcam
• batch recognition and classification
• continuous (image recognition from webcam (polling)
• text to speech
• hardware recognition prototype?

Images to be analysed

put them in the /images directory

images must be under 5mb and in either .jpg or .png format

Installation overview

1. Install Node.js

get it from https://nodejs.org/en/

2. Clone this repository

git clone https://github.com/ajyounguk/img-rekog

3. Install project modules/dependencies

cd img-rekog
npm install

4. Create AWS credentials and add them to the configuration file in the /config directory

You will need a AWS IAM user configured with AmazonRekognitionFullAccess permissions please take care and don't commit your creds back to git if you clone this repo

cd config
cp aws-config-sample.json aws-config.json

edit the aws-config.json file and add your IAM access key (for example, AKIAIOSFODNN7EXAMPLE), the secret access key and your aws account region into the config file

IAM policy requires access to 'rekognition' APIs - e.g. arn:aws:iam::aws:policy/AmazonRekognitionFullAccess

How to run it

run the webserver:

node app.js

How to access the app

point your browser at the local/remoteIP port 3000 to load the HTML form

e.g http://127.0.0.1:3000/