Using AWS Rekognition and Python to Identify Objects / Text in Images
In this video, we'll use AWS Rekognition to identify driver's licenses using machine learning but without having to write any ML code.
Quick Jump: Demo Video
I recently had some client work where we needed to identify if a specific object existed within an image that a user uploaded. From there I discovered AWS Rekognition and here’s a quick introduction and demo of using it.
In the working example we go over how it works in the AWS console and then review a few lines of Python but you can absolutely apply this with any programming language since we’re just calling out to AWS Rekognition using the AWS SDK.
Demo Video
Timestamps
- 0:47 – Demo of using AWS Rekognition in the AWS console
- 1:49 – Uploading custom images to evaluate driver’s licenses
- 4:40 – Extracting specific text out of a driver’s license
- 5:25 – One way to apply this in a real web application
- 7:07 – Writing a little Python script to upload images to AWS Rekognition
- 9:05 – It’s $1 to process 1,000 images or $0.001 dollar per image
- 9:38 – How the Python script works to get the image’s labels
- 11:10 – Extracting text out of an uploaded image with the Python script
- 11:55 – You can also train it with custom rules without writing machine learning code
Reference Links
What are you using AWS Rekognition on? Let me know below.