private static void ThreadProc(object stateInfo)
{
int threadIndex = (int)((object[])stateInfo)[0];
ManualResetEvent doneEvent = (ManualResetEvent)((object[])stateInfo)[1];
string inputFile = (string)((object[])stateInfo)[2];
string outputFile = (string)((object[])stateInfo)[3];
int startPage = (int)((object[])stateInfo)[4];
int endPage = (int)((object[])stateInfo)[5];
try
{
Console.WriteLine("Thread #{0} started with the page range from {1} to {2}.", threadIndex, startPage, endPage);
Stopwatch stopwatch = Stopwatch.StartNew();
// Extract a piece of document
string chunk = string.Format("temp-{0}-{1}", startPage, endPage);
using (DocumentSplitter splitter = new DocumentSplitter("demo", "demo"))
splitter.ExtractPageRange(inputFile, chunk, startPage + 1, endPage + 1);
// Process the piece
using (SearchablePDFMaker searchablePdfMaker = new SearchablePDFMaker("demo", "demo"))
{
searchablePdfMaker.OCRDetectPageRotation = true;
searchablePdfMaker.OCRLanguageDataFolder = @"C:\Program Files\Bytescout PDF Extractor SDK\net4.00\tessdata";
searchablePdfMaker.LoadDocumentFromFile(chunk);
// 300 DPI resolution is recommended.
// Using of higher values will slow down the processing but does not guarantee the higher quality.
searchablePdfMaker.OCRResolution = 300;
searchablePdfMaker.MakePDFSearchable(outputFile);
}
File.Delete(chunk);
Console.WriteLine("Thread #{0} finished in {1}.", threadIndex, stopwatch.Elapsed);
}
finally
{
// Signal the thread is finished
doneEvent.Set();
// Release semaphore
ThreadLimiter.Release();
}
}
private static void ThreadProc(object stateInfo)
{
int threadIndex = (int)((object[])stateInfo)[0];
ManualResetEvent doneEvent = (ManualResetEvent)((object[])stateInfo)[1];
string inputFile = (string)((object[])stateInfo)[2];
string outputFile = (string)((object[])stateInfo)[3];
int startPage = (int)((object[])stateInfo)[4];
int endPage = (int)((object[])stateInfo)[5];
try
{
Console.WriteLine("Thread #{0} started with the page range from {1} to {2}.", threadIndex, startPage, endPage);
Stopwatch stopwatch = Stopwatch.StartNew();
// Extract a piece of document
string chunk = string.Format("temp-{0}-{1}", startPage, endPage);
using (DocumentSplitter splitter = new DocumentSplitter("demo", "demo"))
splitter.ExtractPageRange(inputFile, chunk, startPage + 1, endPage + 1);
/*
* By default, "SearchablePDFMaker" uses one of the standard PDF fonts to apply
* recognized text over the scanned document. Such fonts contain only basic characters
* from ISO-8859-1 charset.
* If you run OCR for one of the languages with characters that are not present in the default
* encoding, you should explicitly specify the font that contains the required characters
* using ".LabelingFont" property.
* If you run the application in Windows with a selected locale that matches OCR language,
* it will be enough to specify the usual font "Arial". But if your app will run in an unknown
* environment (for example, in some virtual machine) you will need to install some full Unicode
* font (e.g. "Arial Unicode MS") and then use it with SearchablePDFMaker:
*
* //searchablePDFMaker.LabelingFont = "Arial Unicode MS";
*/
// Process the piece
using (SearchablePDFMaker searchablePdfMaker = new SearchablePDFMaker("demo", "demo"))
{
searchablePdfMaker.OCRDetectPageRotation = true;
searchablePdfMaker.OCRLanguageDataFolder = @"c:\Program Files\Bytescout PDF Extractor SDK\ocrdata_best\";
searchablePdfMaker.LoadDocumentFromFile(chunk);
// 300 DPI resolution is recommended.
// Using of higher values will slow down the processing but does not guarantee the higher quality.
searchablePdfMaker.OCRResolution = 300;
searchablePdfMaker.MakePDFSearchable(outputFile);
}
File.Delete(chunk);
Console.WriteLine("Thread #{0} finished in {1}.", threadIndex, stopwatch.Elapsed);
}
finally
{
// Signal the thread is finished
doneEvent.Set();
// Release semaphore
ThreadLimiter.Release();
}
}
