Edge Detection

The Document Scanner SDK uses computer vision and machine learning to automatically find the document’s edges in the image and crop it accordingly. For a high-resolution scan, the user should move the camera close to the document, but not so close that its edges move off the screen. This ensures both that only the document itself is captured and also that all of it is visible.

All text in the cropped image can then be recognized using optical character recognition for further processing or to turn the document into a searchable PDF.