For user zones expected to contain either hand-written or machine-printed text, CSDK provides automatic filling method detection.

CSDK leverages a hand-print detection module based on a Convolutional Neural Network (CNN). The module classifies each pixel on the image based on the adjacent pixels and decides whether they belong to a machine-printed or hand-printed text.

To apply the automatic detection, set the filling method for the user zone to FM_AUTO_HAND. When processing the zone, the kRecRecognize function invokes kRecLocateZones, which runs the hand-print detection module on the zone area. If most of the black pixels are classified as a part of handwritten text, kRecLocateZones treats the zone accordingly. Depending on this result the OCR zone's filling method is changed to either FM_HANDPRINT or FM_OMNIFONT, initiating the processing with the appropriate engine. Using FM_AUTO_HAND improves accuracy but requires more processing time, so use it with caution. Due to performance reasons, it is still recommended to use a standard fill method for zones for which the fill method is known in advance..

For better recognition accuracy, consider filtering, which is available for forms as well.