Kaze_Kage - NetrAI

[Boltuzamaki]

Before you start, please follow this format for your issue title:
Kaze_kage - NetrAI

ℹ️ Project information

Please complete all applicable.

• Project Name: NetrAI
• Short Project Description: Helps Visually Impaired People in daily tasks
• Team Name: Kaze_kage
• Team Members: Divyanshu @Boltuzamaki, Abhishek @AbhiDhariwal, Aman @aman28singhal
• Where to find you?: LH19
• Demo Link: https://drive.google.com/drive/folders/1onbVgTgzN6f4ptosoYVI06C2WwgTHOHW?usp=sharing
• Repository Link(s): https://github.com/Boltuzamaki/Kaze_kage
• Prize Categories Enrolled for:
  (Please tick any specific Category that applies)
  • Best use of Microsoft Azure
  • Best Beginner (Your team should be atleast 50% Beginners)
  • Best "Jugaad" - Best Fun/ Silly/ Innovative Hack
  • Best UIPath Hack

🔥 Your Pitch

There are approx 135,000,000 people Visually Impaired in the world.
Who face difficulties in their daily tasks. Problems faced by Visually Impaired include Navigating around the places, Surfing Internet, Using the ATM, Identifying the object and much more.
So, we decided to solve some of their problems with Computer Vision that is available to us.
the key problems we addressed here are:

1. Object identification in the surroundings
   the user just has to point the camera and computer vision will do all of the work. It will read aloud what is in front of them.

The consumer products available in the market are not Braille Friendly which makes it difficult for them.
2. Reading Text
the user just has to point the camera and computer vision will do all of the work. It will read aloud the text.

3. Reading manufacturing and expiry date on consumer products.
   The same has to be done for this too. But there is some additional feature, we will store the expiry date and make a reminder at the time at which the product expires. It will ensure that they don’t use the product after the expiry date.

We have used Azure Cognitive services for Computer vision, Google’s ML vision library for text detection and gTTTS for text to speech
In future we wish to address some more problems.

🔦 Any other specific thing you want to highlight?

Expiry dates are printed in different conventions and on different backgrounds. So, expiry identification is very limited for now.
for object detection, first we have to click a pic, it couldn't be done in real time which would have been better.
The UI is not very user friendly.
Please refer to the powerpoint too.

✅ Checklist

Before you post the issue:

• [x ] You have followed the issue title format.
• [ x] You have mentioned the correct labels.
• [ x] You have provided all the information correctly.