We create and provision software using Amazon Web Services, designed for image analysis and document engineering at scale. Underlying algorithms cover a wide variety of tasks, the following is a quick summary of several REST API methods that are currently available for testing.
Perform OCR using the text-lines method, robust in the presence non-textual objects like borders, obtain bounding box information and horizontal line details as well.
Invert heat map (x,y,z) data with the heat-map method using a traditional color map.
Rapidly extract image outlines, or apply your own custom kernel for image convolution, using the kernel API method.