/// <summary>
/// Processes a file using ResultRenderers.
/// </summary>
/// <param name="renderer"></param>
/// <param name="filename"></param>
private void ProcessImageFile(IResultRenderer renderer, string filename)
{
IEnumerable <string> configs_file = new List <string>()
{
CONFIGS_FILE
};
using (TesseractEngine engine = new TesseractEngine(Datapath, Language, EngineMode, configs_file))
{
var imageName = Path.GetFileNameWithoutExtension(filename);
using (var pixA = LoadPixArray(filename))
{
using (renderer.BeginDocument(imageName))
{
foreach (var pix in pixA)
{
using (var page = engine.Process(pix, imageName))
{
var addedPage = renderer.AddPage(page);
}
}
}
}
}
}
private void ProcessByPage(string pFilePath)
{
string fileNameWithoutExtension = Path.GetFileNameWithoutExtension(pFilePath);
string caminhoPdf = Path.Combine(txbDirSaida.Text, fileNameWithoutExtension); //Caminho onde ira salvar o pdf com OCR sem informar extensao
using (IResultRenderer render = ResultRenderer.CreatePdfRenderer(caminhoPdf, _tessDataPath)) //Caminho pdf e caminho para a fonte do pdf
using (PixArray pages = PixArray.LoadMultiPageTiffFromFile(pFilePath)) //Carrega todas as páginas do tiff
using (render.BeginDocument(fileNameWithoutExtension)) //Cria o pdf
{
List <Task> lstTasksRunning = new List <Task>();
SortedList <int, Page> lstProcessedPages = new SortedList <int, Page>();
int currentPage = 0;
foreach (Pix page in pages)
{
int pageIndex = currentPage;
lstTasksRunning.Add(Task.Factory.StartNew(() =>
{
using (TesseractEngine tesseract = new TesseractEngine(_tessDataPath, "por")) //Caminho da pasta com arquivos de config/ idioma do OCR
{
Page pagina = tesseract.Process(page, Path.GetFileNameWithoutExtension(fileNameWithoutExtension));
lstProcessedPages.Add(pageIndex, pagina); //Processa o arquivo podendo retirar as informacoes de OCR etc.
}
}));
currentPage++;
}
Task.WaitAll(lstTasksRunning.ToArray());
foreach (var pageKeyValue in lstProcessedPages)
{
render.AddPage(pageKeyValue.Value); //Adiciona a pagina
}
}
}
private void ProcessImageFile(IResultRenderer renderer, string filename)
{
var imageName = Path.GetFileNameWithoutExtension(filename);
using (var pixA = ReadImageFileIntoPixArray(filename))
{
int expectedPageNumber = -1;
using (renderer.BeginDocument(imageName))
{
Assert.AreEqual(renderer.PageNumber, expectedPageNumber);
foreach (var pix in pixA)
{
using (var page = _engine.Process(pix, imageName))
{
var addedPage = renderer.AddPage(page);
expectedPageNumber++;
Assert.That(addedPage, Is.True);
Assert.That(renderer.PageNumber, Is.EqualTo(expectedPageNumber));
}
}
}
Assert.That(renderer.PageNumber, Is.EqualTo(expectedPageNumber));
}
}
private void ProcessMultipageTiff(IResultRenderer renderer, string filename)
{
var imageName = Path.GetFileNameWithoutExtension(filename);
using (var pixA = PixArray.LoadMultiPageTiffFromFile(filename))
{
var expectedPageNumber = -1;
using (renderer.BeginDocument(imageName))
{
Assert.AreEqual(renderer.PageNumber, expectedPageNumber);
foreach (var pix in pixA)
{
using (var page = _engine.Process(pix))
{
var addedPage = renderer.AddPage(page);
expectedPageNumber++;
Assert.IsTrue(addedPage);
Assert.AreEqual(renderer.PageNumber, expectedPageNumber);
}
}
}
Assert.AreEqual(renderer.PageNumber, expectedPageNumber);
}
}
public PicklerApplication(
IArgumentParser argumentParser,
IFileLoader fileLoader,
IFeatureExtractor featureExtractor,
ITrxResultsParser trxResultsParser,
IResultRenderer resultRenderer)
{
_argumentParser = argumentParser;
_fileLoader = fileLoader;
_featureExtractor = featureExtractor;
_trxResultsParser = trxResultsParser;
_resultRenderer = resultRenderer;
}
public void Render(IResultRenderer renderer)
{
foreach (var token in _tokens)
{
if (token.Changed)
{
renderer.AppendChanged(token.Text);
}
else
{
renderer.AppendUnchanged(token.Text);
}
}
renderer.End();
}
/// <summary>
/// Processes a file using ResultRenderers.
/// </summary>
/// <param name="renderer"></param>
/// <param name="filename"></param>
private void ProcessImageFile(IResultRenderer renderer, string filename)
{
IEnumerable <string> configs_file = new List <string>()
{
CONFIGS_FILE
};
using (TesseractEngine engine = new TesseractEngine(Datapath, Language, EngineMode, configs_file))
{
ControlParameters(engine);
Tesseract.PageSegMode psm = (PageSegMode)Enum.Parse(typeof(PageSegMode), PageSegMode);
var imageName = Path.GetFileNameWithoutExtension(filename);
using (var pixA = LoadPixArray(filename))
{
using (renderer.BeginDocument(imageName))
{
foreach (var pix in pixA)
{
Pix pixd = null;
try
{
if (Deskew)
{
pixd = pix.Deskew(new ScewSweep(range: 45), Pix.DefaultBinarySearchReduction, Pix.DefaultBinaryThreshold, out Scew scew);
}
using (var page = engine.Process(pixd ?? pix, imageName, psm))
{
var addedPage = renderer.AddPage(page);
}
}
finally
{
if (pixd != null)
{
((IDisposable)pixd).Dispose();
}
}
}
}
}
}
}
private void ProcessFile(IResultRenderer renderer, string filename)
{
var imageName = Path.GetFileNameWithoutExtension(filename);
using (var pix = Pix.LoadFromFile(filename)) {
using (renderer.BeginDocument(imageName)) {
Assert.AreEqual(renderer.PageNumber, -1);
using (var page = _engine.Process(pix, imageName)) {
var addedPage = renderer.AddPage(page);
Assert.That(addedPage, Is.True);
Assert.That(renderer.PageNumber, Is.EqualTo(0));
}
}
Assert.AreEqual(renderer.PageNumber, 0);
}
}
private void ProcessByFile(string pFilePath)
{
using (TesseractEngine tesseract = new TesseractEngine(_tessDataPath, "por")) //Caminho da pasta com arquivos de config/ idioma do OCR
{
string fileNameWithoutExtension = Path.GetFileNameWithoutExtension(pFilePath);
string caminhoPdf = Path.Combine(txbDirSaida.Text, fileNameWithoutExtension); //Caminho onde ira salvar o pdf com OCR sem informar extensao
using (IResultRenderer render = ResultRenderer.CreatePdfRenderer(caminhoPdf, _tessDataPath)) //Caminho pdf e caminho para a fonte do pdf
using (PixArray pages = PixArray.LoadMultiPageTiffFromFile(pFilePath)) //Carrega todas as páginas do tiff
using (render.BeginDocument(fileNameWithoutExtension)) //Cria o pdf
{
foreach (Pix page in pages)
{
using (page)
using (Page processedPage = tesseract.Process(page, Path.GetFileNameWithoutExtension(fileNameWithoutExtension))) //Processa o arquivo podendo retirar as informacoes de OCR etc.
render.AddPage(processedPage); //Adiciona a pagina
}
}
}
}
private static void ProcessImageFile(IResultRenderer renderer, string filename)
{
var fileName = Path.GetFileNameWithoutExtension(filename);
using (var engine = new TesseractEngine(TesseractData, "eng", EngineMode.Default))
{
using (var pixA = ReadImageFileIntoPixArray(filename))
{
int expectedPageNumber = -1;
foreach (var pix in pixA)
{
using (var page = engine.Process(pix, fileName))
{
using (renderer.BeginDocument("multipage_tiff_example"))
{
var addedPage = renderer.AddPage(page);
expectedPageNumber++;
}
}
}
}
}
}
/// <summary>
/// Creates renderers for specified output formats.
/// </summary>
/// <param name="outputbase"></param>
/// <param name="dataPath">The directory containing the pdf font data, normally same as your tessdata directory.</param>
/// <param name="outputFormats"></param>
/// <returns></returns>
public static IResultRenderer CreateRenderers(string outputbase, string dataPath,
List <RenderedFormat> outputFormats)
{
IResultRenderer renderer = null;
foreach (var format in outputFormats)
{
switch (format)
{
case RenderedFormat.TEXT:
if (renderer == null)
{
renderer = CreateTextRenderer(outputbase);
}
else
{
TessApi.Native.ResultRendererInsert(((ResultRenderer)renderer).Handle,
new TextResultRenderer(outputbase).Handle);
}
break;
case RenderedFormat.HOCR:
if (renderer == null)
{
renderer = CreateHOcrRenderer(outputbase);
}
else
{
TessApi.Native.ResultRendererInsert(((ResultRenderer)renderer).Handle,
new HOcrResultRenderer(outputbase).Handle);
}
break;
case RenderedFormat.PDF:
if (renderer == null)
{
renderer = CreatePdfRenderer(outputbase, dataPath, false);
}
else
{
TessApi.Native.ResultRendererInsert(((ResultRenderer)renderer).Handle,
new PdfResultRenderer(outputbase, dataPath, false).Handle);
}
break;
case RenderedFormat.BOX:
if (renderer == null)
{
renderer = CreateBoxRenderer(outputbase);
}
else
{
TessApi.Native.ResultRendererInsert(((ResultRenderer)renderer).Handle,
new BoxResultRenderer(outputbase).Handle);
}
break;
case RenderedFormat.UNLV:
if (renderer == null)
{
renderer = CreateUnlvRenderer(outputbase);
}
else
{
TessApi.Native.ResultRendererInsert(((ResultRenderer)renderer).Handle,
new UnlvResultRenderer(outputbase).Handle);
}
break;
}
}
return(renderer);
}
public static void Main(string[] args)
{
var testImagePath = "./closeup_scan.png";//"./phototest.tif";
if (args.Length > 0)
{
testImagePath = args[0];
}
try
{
using (IResultRenderer renderer = ResultRenderer.CreatePdfRenderer(@"./output", @"./tessdata"))
{
using (renderer.BeginDocument("PDF Test"))
{
using (TesseractEngine engine = new TesseractEngine(@"./tessdata", "eng", EngineMode.TesseractAndCube))
{
var list = Directory.GetFiles(@"images");
foreach (var item in list)
{
using (var tifFile = new Bitmap(item))
{
//using (var img = PixConverter.ToPix(tifFile))
//{
using (var page = engine.Process(tifFile, "test"))
{
renderer.AddPage(page);
}
//}
}
}
}
}
}
//using (var engine = new TesseractEngine(@"./tessdata", "eng", EngineMode.Default))
//{
// using (var img = Pix.LoadFromFile(testImagePath))
// {
// using (var page = engine.Process(img))
// {
// var text = page.GetText();
// Console.WriteLine("Mean confidence: {0}", page.GetMeanConfidence());
// Console.WriteLine("Text (GetText): \r\n{0}", text);
// Console.WriteLine("Text (iterator):");
// using (var iter = page.GetIterator())
// {
// iter.Begin();
// do
// {
// do
// {
// do
// {
// do
// {
// if (iter.IsAtBeginningOf(PageIteratorLevel.Block))
// {
// Console.WriteLine("<BLOCK>");
// }
// Console.Write(iter.GetText(PageIteratorLevel.Word));
// Console.Write(" ");
// if (iter.IsAtFinalOf(PageIteratorLevel.TextLine, PageIteratorLevel.Word))
// {
// Console.WriteLine();
// }
// } while (iter.Next(PageIteratorLevel.TextLine, PageIteratorLevel.Word));
// if (iter.IsAtFinalOf(PageIteratorLevel.Para, PageIteratorLevel.TextLine))
// {
// Console.WriteLine();
// }
// } while (iter.Next(PageIteratorLevel.Para, PageIteratorLevel.TextLine));
// } while (iter.Next(PageIteratorLevel.Block, PageIteratorLevel.Para));
// } while (iter.Next(PageIteratorLevel.Block));
// }
// }
// }
//}
}
catch (Exception e)
{
Trace.TraceError(e.ToString());
Console.WriteLine("Unexpected Error: " + e.Message);
Console.WriteLine("Details: ");
Console.WriteLine(e.ToString());
}
Console.Write("Press any key to continue . . . ");
Console.ReadKey(true);
}
