Not every AI problem needs deep learning.
The number of posts on custom optical character recognition with machine learning is quite small at the point of writing this post. The common ones you will probably see make use of deep learning or the tesseract library. In this post, I will be explaining how to train your own custom optical character recognizer using machine learning. The method we’ll be looking at won’t require much data but performs quite wonderfully. I used this when building a prototype ALPR for Nigerian License plates. So let’s dive in.
First of all, you have to…
Have you ever wanted to deploy your opencv web application on the web and show it off to the world? This is a great place for you to start!
In this article, I’ll walk you through deploying your opencv web app on Heroku. We will first build a MobileNet SSD object detector using Flask and then deploy it on Heroku. The codes for this project can be found here.
In the wake of this pandemic I wanted to make myself useful and build something simple and at the same time put my web scrapping skills to use. So I decided to build an API off NCDC’s website (https://covid19.ncdc.gov.ng/) and be able to use it to build an information bot using Twilio’s whatsapp sandbox. Well this write up is just focused on building the API used in the information bot and as you can guess it was built using python.
Python Developer, focusing on artificial intelligence, machine learning and web application development |https://www.linkedin.com/in/oluwaseun-ilori/