To extract the textual information from an image, Optical Character Recognition (OCR) technique is utilized.
OCR is a technology that recognizes the text characters from an image and converts it into machine-readable text.
It works by analyzing the pixel pattern of the input image and matches it with a known set of character patterns
to recognize the characters.
Our API provides two AI-powered OCR engines to extract text from your images.
The first OCR engine is named
scan and is optimized to extract text from clean document pages
captured using a scanner or camera. The accuracy of the output is directly related to the quality of the input
image, and the optimal conditions for this engine are 300dpi resolution, black text on a white background,
and a character height of around 20pts.
However, if these optimal conditions cannot be met, or if the expected results are not achieved, we offer
a second OCR engine called
picture. This engine is designed to extract text from any source image and is
capable of recognizing text even in noisy images. Please note that using the "picture" OCR engine may result
in the loss of formatting in paragraphs. However, the textual information will still be recognized with a high degree
Please note that the
language option can be set to specify which language the text is in. This can lead to a better conversion result.