Home

Tesserwrap

Tesserwrap is a Python binding to the Tesseract-OCR API (provided by libtesseract_api). Currently Tesserwrap is Alpha quality with no warranty. The goal is to create a simple way to OCR images in Python without calling the Tesseract application directly and creating temporary files.

• Installation
• Examples

Currently Supports:

• OCRing an image in one-shot.
• Setting an image and bounding rectangle.
• Setting segmentation modes.
• Large pictures. (All of the test images I use are 600dpi. I have yet to test with lower resolution images)

Future Support/Ideas:

• Leptonica image format conversion. (Leptonica PIX is the native format for Tesseract)
• Processing OCR boundaries in C++ and returning an array of strings. (This may give a slight speed boost)
• Thread-safety. Currently I would NOT try to access this object from multiple threads without using a Semaphore.