Rekognition Managed Service In AWS


In this article, we will learn about one of the AWS-managed services called Rekognition service. This service is a comprehensive object recognition service that can also recognize text in images.

About Rekognition Managed Service

Amazon Rekognition can detect text in images and videos. It can then convert the detected text into machine-readable text. You can use machine-readable text detection in images to implement solutions. This solution is supposed to be used in the following scenarios

  • Visual Search
  • Filtering the text
  • Searching content in video clips
  • and many more 

This service is designed to detect words in a few languages, such as English, Arabic, Russian, German, French, Italian, Portuguese, and Spanish.

Why Rekognition is more powerful than TextReader service 

Amazon Textract is a newer AWS service created as a purpose-built solution to the problem of OCR (optical character recognition) in images of documents and PDFs. While Rekognition is a more generalizable computer vision service, Textract has many more OCR-oriented tuning parameters to optimize the process of accurately and effectively extracting text.

Out of the box, if all you are trying to do is detect text and the relevant metadata (coordinates, angle, confidence value), the Rekognition DetectText method will likely perform similarly to the equivalent analyze_document method in Textract. However, Textract offers further semantic structuring that helps with text curation/formatting that abstracts other forms of post-processing that the developer would traditionally need to write themselves.

Lastly, when comparing the costs of the two Detect Text methods, Textract costs a bit more ($1.50/1k images) than Rekognition ($1.00/1k images).

Typical Use Case Architecture using Rekognition managed Service

Rekognition Managed Service in AWS

In the following architecture diagram, our use case is as follows,

  • Take a snapshot of the whole car
    • Read the FastTag Number or BarCode
    • Reading the Vehicle Registration number 
  • Extract both the number in a machine-readable format
  • Validate the following scenarios
    • Validate the Vehicle RegNumber with the Vehicle Model. Is this correct with the help of RTO API?
    • Validate the FastTag is not blacklisted
    • Is this vehicle exempted such as an armed force vehicle or VIP vehicle
    • Apply the fee logic and pass it to the payment API to consume it
    • Finally, respond to Kafka from the event bridge with the appropriate status
  • Finally, Toll may open or display an appropriate message for further action. 

I hope this article explains why we use Rekognition and TextReader managed services and their typical use cases. 

Similar Articles