Photerpret

Ingredients and nutrition info for a KitKit bar

What it does

Photerpret uses Microsoft Azure and IBM Watson to scan through folders of product images, parsing for important text, such as nutrition info, ingredients, and more!

How we built it

We used Microsoft Azure's Computer Vision API to identify text and then customized the text interpretation based on image type. IBM Watson was used for secondary analysis of picture details.

Challenges we ran into

Choosing what API to use for text recognition and analysis was difficult. The output from Azure was formatted in highly nested lists and dictionaries, making it extremely tough to gather the appropriate data. Figuring out how to pass images using JSON was a challenge as well.

Accomplishments that we're proud of

The algorithms used to manipulate the output into meaningful data structures. Being able to present a complete product after two days of development.

What we learned

Deciding which API to use for image analysis should've been done carefully. Caffeine is amazing.

What's next for Photerpret

Web API integration

Built With

Submitted to

QHacks 2018

Created by

I worked a lot on creating own api. It was my first time ever going to a hackathon and attempting to make my own api so there were a lot of problems but I believe I learned a lot from this experience.

Daniel Tavernese
I worked mainly on extracting JSON information and packaging it into nested lists and dictionaries in python. This project encouraged me to explore Flask and Django.

Private user
AndyHFW

Updates

AndyHFW started this project — Feb 04, 2018 06:41 AM EST

Leave feedback in the comments!

Log in or sign up for Devpost to join the conversation.