Posted On: Nov 17, 2021

Amazon Rekognition can detect and read text in an image, and return bounding boxes for each word found. Starting today, Amazon Rekognition supports text detection in images in 7 new languages - Arabic, Russian, German, French, Italian, Portuguese and Spanish. Amazon Rekognition automatically detects and extracts text in images in all supported languages, without requiring a language parameter. In addition, Amazon Rekognition delivers higher overall accuracy, with improvements for vertical and curved text in images.

Customers can use text detection in images for multiple use cases. First, text detection can support content moderation workflows, where detected text can be checked against a list of inappropriate or unwanted words and phrases. Second, customers can use the detected text bounding box area to redact personally identifiable information (PII). Third, text detection can be used to understand how particular words, text placement, and text size impact marketing campaign performance. Fourth, customers can use text detection to easily search for image or video assets with specific keywords or captions in the Digital Asset Management (DAM) system. Additionally, text detection supports mapping and automotive applications, to read text on road and street signs, and in public safety and transportation. Following is a quote from OLX Global on how they are using Amazon Rekognition for text detection in images:

As a leader in the classified marketplace sector and to foster a safe, inclusive and vibrant buying and selling community, it is paramount that we make sure that all products listed on our platforms comply with our rules for product display and authenticity. To do that, among other aspects of the ads, we have placed focus on analyzing the non-organic text featured on images uploaded by our users. We tested Amazon Rekognition’s text detection functionality for this purpose and found that it was highly accurate and augmented our in-house violations detection systems, helping us improve our moderation workflows. Using Amazon Rekognition for text detection we were able to flag 350,000 policy violations last year. It has also helped us save significant amounts in development costs and has allowed re-focus data science time on other projects. We are very excited about the upcoming text model update as it will even further expand our capabilities for text analysis.”- Jaroslaw Szymczak, Data Science Manager, OLX Group 

These improvements are now available in all AWS Regions supported by Amazon Rekognition. To get started, please visit the Rekognition Console or download the latest AWS SDK to start building applications. To learn more about text detection with Amazon Rekognition, please refer to our documentation.