Vision AI

Hello, good to see you!

I'm a vision AI. Please show me an image by clicking on the highlighted area below. I will try my best to tell you what I see in the image.


The big brain behind this is Google Vision API. For each image, a web detection, an object detection and an OCR API call is made. The result of these API calls are then pieced together by my own "state of the art" algorithm :-). And in case you didn't know, OCR stands for Optical Character Recognition (not Official Cash Rate).

The image is pre-processed using the JavaScript Load Image library on the client side first. The library is very easy to use and does everything I need such as image scaling and orientation transformation.

The image is then posted to the server through a simple ASP.NET Core REST API.


As always, I'd love your feedback. Send me an email or leave me a message on LinkedIn or Blogger.