Posted On: Aug 6, 2018

Amazon Rekognition has launched improvements for Text-in-Image to increase the accuracy of text detection and to provide expanded support for text rotated -90 to +90 degrees from the horizontal axis.

Amazon Rekognition is a deep learning-based computer vision service that makes it easy to add image and video analysis to your applications. With Amazon Rekognition Text-in-Image, you can detect and extract text in social images to provide additional context, as well as accurately detect and extract text from real-world images, such as names and numbers on road signs or words in advertisements. With these improvements, customers are now able to detect text with even greater accuracy and analyze much more challenging images such as skewed and rotated text.

Amazon Rekognition’s API enables you to quickly add sophisticated deep learning-based visual search and image classification to your applications. With Rekognition, you only pay for the images and video you analyze and the face metadata you store. There are no minimum fees or upfront commitments.

Amazon Rekognition is also currently available in US East (Northern Virginia), US West (Oregon), EU (Ireland), and Asia Pacific (Sydney and Tokyo) regions. To get started, simply log in to the Rekognition Console to try the service with sample images or your own images.