ABBYY FineReader Engine 9.0 CLI for Linux
Recognition Keys

www.ocr4linux.com / Documentation Home /Step 2 Recognition: Recognition

 

Key Parameters Default Description

-rbm

--balancedMode

no

 

The recognition will run in balanced mode (an intermediate mode between full and fast modes, which can be activated with the help of the -fm key).

-rcm

--caseRecognitionMode

AutoCase
Automatically detects the case of letters and keeps it in the output text.

SmallCase
The recognised text will be set in lowercase letters.

CapitalCase
The recognised text will be set in capitals.

AutoCase

Specifies the mode of letter case recognition.

-rcc

--cellsCount

the number of character cells for a recognised block

1

Specifies the number of character cells for a recognised block. This key is valid only for the handprint recognition. It has a sense only for the field marking types (-rfmt key) that imply splitting the text in cells.

-rcjktd

--cjkTextDirection*

Horizontal
The text to be recognised is arranged horizontally.

Vertical
The text to be recognised is arranged vertically. Characters are written one below the other, top to bottom.

Autodetect
The direction of the text is detected automatically.

Autodetect

Specifies the direction of the text to be recognised. This property is valid only for the hieroglyphic languages.

-rdss

--disableSecondStageRecognizer

no

 

No second-stage recognition will be performed.

-rehl

--errorHiliteLevel

None
Do not highlight uncertain characters at all.

Scanty
Highlight only very uncertain characters.

Standard
Sets standard highlight level.

Thorough
Highlight each character that is uncertain.

AllText
Highlight all characters in the text.

Standard

Specifies the level at which the ICharParams::IsSuspicious property is set to TRUE for a recognised character.

-recc

--exactConfidenceCalculation*

no

 

Character confidence will be defined more accurately, but recognition speed may get slower.

-fm

--fastMode

no

 

Provides 2-2,5 times faster recognition speed at the cost of moderate increased error rate (1,5-2 times more errors). For text printed with good quality  ABBYY FineReader Engine makes 1-2 errors per page on average. The increase of errors can be easily  eliminated  by using various scenarios, e.g. the full text indexing with "fuzzy" search, preliminary recognition, etc.

-rfmt

--fieldMarkingType

SimpleText
UnderlinedText
TextInFrame
GreyBoxes
CharBoxSeries
SimpleComb
CombInFrame
PartitionedFrame

SimpleText

This property specifies the type of marking around letters (for example, underline, frame, box and etc.). This property is valid only for the handprint recognition.

-tol

--oneLinePerTextCell

no

 

Presumes that the text in the block to which the current RecognizerParams object corresponds contains no more than one string.

-rowl

--oneWordPerLine

no

 

Presumes that no text line may contain more than one word, so the lines of text will be recognised as a single word.

-rptt

--possibleTextTypes

collection of parameters of the -prn key

Normal

Presumes that the text to recognize is of one of the types the collection contains.

-rphp

--prohibitHyphenation

no

 

Prohibits recognition of hyphenation from line to line.

-rpihp

--prohibitInterblockHyphenation

no

 

Presumes that text from one block can't be carried over to the next block.

-rpi

--prohibitItalic

no

 

Prohibits recognition of letters printed with italic-styled font.

-rpsub

--prohibitSubscript

no

 

Prohibits recognition of subscript letters.

-rpsup

--prohibitSuperscript

no

 

Prohibits recognition of superscript letters.

-rscrv

--saveCharacterRecognitionVariants*

no

 

Specifies whether the variants of a character's recognition are saved.

-rswrv

--saveWordRecognitionVariants*

no

 

Specifies whether the variants of a word's recognition are saved.

-rl

 --recognitionLanguage

predefined languages and/or user languages

English

Sets the recognition language(s) which should be present in the list of the predefined ABBYY FineReader Engine languages and/or in the /UserLanguages directory.

-prn

--printType

Normal
Corresponds to a common typographic text.

Typewriter
Presumes that the text on the recognised image is typed on a typewriter.

Matrix
Presumes that the text on the recognised image is printed on a dot matrix printer.

OCR_A
Corresponds to a monospaced font designed for OCR. Widely used by banks, credit card companies and similar businesses.

OCR_B
Corresponds to a font designed for OCR.

MICR_E13B
Corresponds to a special set of numeric characters printed with special magnetic inks. MICR (Magnetic Ink Character Recognition) characters are found in a variety of places, including personal checks.

MICR_CMC7
Corresponds to a special MICR barcode font.

Gothic
Presumes that the text on the recognized image is printed with the Gothic type.

ToBeDetected
Automatically detects the type of the text.

Normal

Sets the text type of recognised document.

Note: If you set print type to ToBeDetected, please consider using -rptt key with print types that may appear in your document.

-upf

--userPatternsFile

user patterns file name

no

This key sets the path to the user patterns file used for recognition.

Note: User patterns can be created in Windows version of ABBYY FineReader Engine 9.0 or ABBYY FineReader 9.0.

-dnubp

--doNotUseBuiltInPatterns

no

 

When -dnubp key is not used, ABBYY FineReader Engine will use its own built-in patterns for recognition. Patterns are files establishing relationship between character image and character itself.

You may want to use this key when you do not want to use standard built-in ABBYY FineReader Engine patterns for character recognition, but user patterns only. This may be useful for recognition of text typed with decorative or non-standard fonts. In this case it is better not to use ABBYY FineReader Engine built-in patterns, but use your own user-defined patterns trained for these fonts.

Please note that when -dnubp key is used, the path to the user-defined patterns file should be specified with the -upf key as only information stored in this file will be used. If the path to the user patterns file is not specified, used -dnubp key is ignored so built-in patterns are used for recognition.

 

User Languages

In the ABBYY CLI OCR 9.0 installation directory (that is /opt/ABBYYOCR9 by default) there is the /UserLanguages subdirectory where you can put user languages created in ABBYY FineReader 9.0.
ABBYY FineReader 9.0 with its user interface provides relatively simple way to create custom recognition languages. The procedure of creating and importing recognition languages is as follows:

 

* XML export with extended character information, 2D barcode recognition and OCR for Asian languages (Chinese Korean Japanese) are not included. If these features are required, please contact your ABBYY's sales representative for a special offer.

Note. Full keys are marked by italic.

See also

ABBYYOCR CLI for Linux - Documentation Home