What is online OCR?

Optical Character Recognition (OCR) is a technology that enables text recognition and conversion of digital images (scans and photos) and PDFs to text ready for editing, quoting, search, and archiving. OCR software or an Online OCR service converts scans, document images taken with a digital camera or smartphone, as well as PDF files to searchable and editable formats.

How does ABBYY FineReader Online OCR work?

Before ABBYY FineReader Online converts an image with optical character recognition it applies special pre-processing functions to correct image distortions (page orientation, skews, speckles, ISO noise, etc.).

The service starts the image recognition process with an analysis of the document structure. It divides the page into elements such as blocks of text, tables and images. Then it detects lines, individual words and letters. Each ‘character picture’ is compared to a set of patterns and the service defines the highest probability of a certain letter. Supported by sophisticated algorithms and ABBYY’s Adaptive Document Recognition Technology (ADRT®) FineReader Online determines the position of each letter in a word, each word in a line as well as the properties of all pages and their relation to each other. The layout of each page with all formatting elements such as paragraphs, tables or columns can thus be exactly reproduced. The logical structure of the whole document, including headers and footers, footnotes, page numbers and table of contents can be reconstructed.

In addition, FineReader Online provides dictionary support for 37 of 42 available OCR languages. It compares detected words with the words in one or several of the dictionaries, which ensures an even higher recognition accuracy.

After the recognition process is finished, the recreated document is exported, i.e. saved in one of the common formats such as DOC/DOCX, XLS/XLSX, ODT, etc. Texts thus obtained can be further edited in special processing applications, quoted, and saved for sharing or archiving.