Insol Group

Recognition of documents

Recognition of documents

Usually working with the information in electronic form we use computer text editors. Often there is a problem of conversion of information from hard-copy form in an electronic format of some text or tabular editors, more often these are Word and Excel editors.

There are two ways out of this situation. The first is to write the information from the hard-copy form manually, the second is to try to automate this process. For this purpose, first of all, it is necessary to scan necessary documents, then to recognize the text using special programs, and finally, in order to get a qualitative result, the document has to be edited and conversed into a convenient format (in Word format or which is better in PDF).

Certainly having certain skills you can do it by yourself, but it is better to use our company’s services because:
— We use the professional equipment and high-speed scanners.
— We already have licensed software and employees who know how to work with it.
— Among our employees there are professional editors for whom it is easy to input formulas, to insert diagrams, etc. The main thing is to make it all quickly and qualitatively.
— You will expend your energies and will spend your time and still the result can disappoint you. And we will make your work quickly and qualitatively. You can check it up by means of a free trial order service.

Scanning of documents, recognition of the text and the further text handling conversion of the scanned documents into the text services:

1. Scanning of the hard-copy form. Scanning is usually made in black-and-white mode with the resolution 300dpi. If there is a necessity of saving any illustrations, diagrams or other images then we make a monochrome (grayscale) scanning or colour.

2. Recognition of a document’s structure. By means of the professional software we make recognition of a document’s and text’s structure. Especially it should me mentioned that automatic structure determination that exists in any software is not perfect and it is necessary to check its correctness and to correct structure.

3. Recognition of the document’s text. After the previous actions, by means of the special software we make text recognition. At this stage it’s really important to establish recognition parameters correctly to reduce error possibility to a minimum, because quantity of errors depends directly on quality of the document, size and contrast, mutual placing of elements of the text.

4th stage. Check of correctness of the text’s recognition. By special equipment and software we make visual check on conformity of poorly distinguished symbols.

5th stage. Check of spelling and syntactic errors. On the third stage it is usually impossible to get rid of all the errors, that’s why it is necessary to check spelling by means of special spelling systems.

6th stage. Formatting of the document. At this stage the final document is created. Uniform font and paragraphs settings are established. Placing and formation of tables is made. Also at this stage we make additional manual input of the text, tables and formulas. At this stage it is possible to achieve almost full similarity of the document with its paper source, including layout and fonts.
Cost of works is defined by quantity of stages which are necessary to use for scanning and document processing. The more stages there will be involved, the more qualitative and exact document you get.

Of course you may order only three or four stages, and the rest of the work you can make by yourself, but generally it will take a lot of man-hours.

Services prices

Automatic recognition — 1 rouble
Recognition with manual setting of recognition zones (text, text and images) — 2 roubles
Recognition with manual setting of recognition zones (tables) — 3 roubles.
Correcting and compliance test (common text) — 2 roubles.
Correcting and compliance test (complex text) — 4 roubles.
Correcting and compliance test (common table) — 4 roubles per table
Correcting and compliance test (complex table) — 8 roubles per table
Image formation — 2 roubles per image.

Types of pages:
Common text — Pages containing only uniform text without difficult design.

Complex text — Page which besides the text contains also different signs, application of various formatting and text design.

Common table —Table consisting of 1-12 lines, without merging of cells and without cells with differently directed text.
Complex table —Full-page table with merged cells and differently directed text. Also not cellated tables.

Image — Graphic element that usually has a signature.

Formula —Formula consisting of difficult symbols and elements that could be displayed by means of special editors for formulas.