Class ResultIterator

Inheritance
System.Object
DisposableBase
PageIterator
ResultIterator
Implements
System.IDisposable
Inherited Members
Namespace: DynamicTesseract
Assembly: IronOcr.dll
Syntax
public sealed class ResultIterator : PageIterator

ResultIterator is what code reads from when it needs both the recognized text and where that text came from. It extends the page traversal of PageIterator with the recognition output Tesseract attaches to each element, so a single walk yields the words, their confidence, and their layout together rather than in separate passes.

A ResultIterator is produced from a recognized page and inherits the cursor mechanics of its base, Begin to reset and Next to advance by a PageIteratorLevel, while adding the result-side reads on top. Because it is sealed and disposable, wrap it in using or dispose it once the walk is done. The geometry methods of the base, such as TryGetBoundingBox, remain available, so position and text are read at the same cursor stop.

GetText returns the recognized text at a chosen PageIteratorLevel, from a single symbol up to a whole block, and GetConfidence reports how certain the engine is about that element so low-confidence results can be flagged or re-read. The word-level reads describe each word in context: GetWordFontAttributes reports its font, GetWordRecognitionLanguage the language the engine matched, and GetWordIsFromDictionary and GetWordIsNumeric flag dictionary words and numeric tokens, which helps when validating fields such as totals or codes. At the symbol level, GetSymbolIsSuperscript, GetSymbolIsSubscript, and GetSymbolIsDropcap distinguish special characters, while GetChoiceIterator exposes the alternative candidates the engine considered for the current symbol when a result needs review.

using DynamicTesseract;

void DumpWords(ResultIterator iterator)
{
    iterator.Begin();
    do
    {
        string word = iterator.GetText(PageIteratorLevel.Word);
        float confidence = iterator.GetConfidence(PageIteratorLevel.Word);
        Console.WriteLine($"{word} ({confidence:F1}%)");
    }
    while (iterator.Next(PageIteratorLevel.Word));
}

The OCR results how-to covers reading recognized text, the results objects example shows the result model, and the result confidence how-to explains the confidence scores.

Methods

GetChoiceIterator()

Declaration
public ChoiceIterator GetChoiceIterator()
Returns
Type Description
ChoiceIterator

GetConfidence(PageIteratorLevel)

Declaration
public float GetConfidence(PageIteratorLevel level)
Parameters
Type Name Description
PageIteratorLevel level
Returns
Type Description
System.Single

GetSymbolIsDropcap()

Declaration
public bool GetSymbolIsDropcap()
Returns
Type Description
System.Boolean

GetSymbolIsSubscript()

Declaration
public bool GetSymbolIsSubscript()
Returns
Type Description
System.Boolean

GetSymbolIsSuperscript()

Declaration
public bool GetSymbolIsSuperscript()
Returns
Type Description
System.Boolean

GetText(PageIteratorLevel)

Declaration
public string GetText(PageIteratorLevel level)
Parameters
Type Name Description
PageIteratorLevel level
Returns
Type Description
System.String

GetWordFontAttributes()

Declaration
public FontAttributes GetWordFontAttributes()
Returns
Type Description
FontAttributes

GetWordIsFromDictionary()

Declaration
public bool GetWordIsFromDictionary()
Returns
Type Description
System.Boolean

GetWordIsNumeric()

Declaration
public bool GetWordIsNumeric()
Returns
Type Description
System.Boolean

GetWordRecognitionLanguage()

Declaration
public string GetWordRecognitionLanguage()
Returns
Type Description
System.String

Implements

System.IDisposable

Inherited members

page
handle
Finalize()
Begin()
Next(PageIteratorLevel)
Next(PageIteratorLevel, PageIteratorLevel)
IsAtBeginningOf(PageIteratorLevel)
IsAtFinalOf(PageIteratorLevel, PageIteratorLevel)
TryGetBoundingBox(PageIteratorLevel, Rect)
TryGetBaseline(PageIteratorLevel, Rect)
GetProperties()
Dispose(Boolean)
BlockType
Dispose()
VerifyNotDisposed()
IsDisposed