Auto Crop Image

The Document Scanner SDK’s algorithms automatically trigger the scan as soon as the device’s camera is in a good position above the document. Computer vision and machine learning is then used to look for hard edges in the image, locate the document boundary, apply perspective correction, and crop it accordingly. No trace of the image’s background remains.

This ensures that all the text in the input image can be recognized using optical character recognition to turn the document into a searchable PDF. You can also apply this feature to files from the device’s image gallery.