public virtual void ComparePdfA3uRGBSpanishJPG()
{
String testName = "comparePdfA3uRGBSpanishJPG";
String filename = "spanish_01";
String expectedPdfPath = TEST_DOCUMENTS_DIRECTORY + filename + "_a3u.pdf";
String resultPdfPath = GetTargetDirectory() + filename + "_" + testName + "_a3u.pdf";
Tesseract4OcrEngineProperties properties = new Tesseract4OcrEngineProperties(tesseractReader.GetTesseract4OcrEngineProperties
());
properties.SetPathToTessData(GetTessDataDirectory());
properties.SetLanguages(JavaCollectionsUtil.SingletonList <String>("spa"));
tesseractReader.SetTesseract4OcrEngineProperties(properties);
OcrPdfCreatorProperties ocrPdfCreatorProperties = new OcrPdfCreatorProperties();
ocrPdfCreatorProperties.SetPdfLang("en-US");
ocrPdfCreatorProperties.SetTitle("");
ocrPdfCreatorProperties.SetTextColor(DeviceRgb.BLACK);
OcrPdfCreator ocrPdfCreator = new OcrPdfCreator(tesseractReader, ocrPdfCreatorProperties);
PdfDocument doc = ocrPdfCreator.CreatePdfA(JavaCollectionsUtil.SingletonList <FileInfo>(new FileInfo(TEST_IMAGES_DIRECTORY
+ filename + ".jpg")), GetPdfWriter(resultPdfPath), GetRGBPdfOutputIntent());
NUnit.Framework.Assert.IsNotNull(doc);
doc.Close();
NUnit.Framework.Assert.IsNull(new CompareTool().CompareByContent(resultPdfPath, expectedPdfPath, GetTargetDirectory
(), "diff_"));
}
public virtual void ComparePdfA3uCMYKColorSpaceJPG()
{
String testName = "comparePdfA3uCMYKColorSpaceJPG";
String filename = "numbers_01";
String expectedPdfPath = TEST_DOCUMENTS_DIRECTORY + filename + "_a3u.pdf";
String resultPdfPath = GetTargetDirectory() + filename + "_" + testName + "_a3u.pdf";
try {
OcrPdfCreatorProperties ocrPdfCreatorProperties = new OcrPdfCreatorProperties();
ocrPdfCreatorProperties.SetPdfLang("en-US");
ocrPdfCreatorProperties.SetTitle("");
OcrPdfCreator ocrPdfCreator = new OcrPdfCreator(tesseractReader, ocrPdfCreatorProperties);
tesseractReader.SetTesseract4OcrEngineProperties(tesseractReader.GetTesseract4OcrEngineProperties().SetTextPositioning
(TextPositioning.BY_WORDS));
NUnit.Framework.Assert.AreEqual(tesseractReader, ocrPdfCreator.GetOcrEngine());
ocrPdfCreator.SetOcrEngine(tesseractReader);
PdfDocument doc = ocrPdfCreator.CreatePdfA(JavaCollectionsUtil.SingletonList <FileInfo>(new FileInfo(TEST_IMAGES_DIRECTORY
+ filename + ".jpg")), GetPdfWriter(resultPdfPath), GetCMYKPdfOutputIntent());
NUnit.Framework.Assert.IsNotNull(doc);
doc.Close();
NUnit.Framework.Assert.IsNull(new CompareTool().CompareByContent(resultPdfPath, expectedPdfPath, GetTargetDirectory
(), "diff_"));
}
finally {
NUnit.Framework.Assert.AreEqual(TextPositioning.BY_WORDS, tesseractReader.GetTesseract4OcrEngineProperties
().GetTextPositioning());
tesseractReader.SetTesseract4OcrEngineProperties(tesseractReader.GetTesseract4OcrEngineProperties().SetTextPositioning
(TextPositioning.BY_LINES));
}
}
public virtual void TestTextFromPdfLayersFromMultiPagePdf()
{
String testName = "testTextFromPdfLayersFromMultiPagePdf";
String pdfPath = GetTargetDirectory() + testName + ".pdf";
IList <FileInfo> files = JavaUtil.ArraysAsList(new FileInfo(TEST_IMAGES_DIRECTORY + "german_01.jpg"), new FileInfo
(TEST_IMAGES_DIRECTORY + "tèst/noisy_01.png"), new FileInfo(TEST_IMAGES_DIRECTORY + "nümbérs.jpg"), new
FileInfo(TEST_IMAGES_DIRECTORY + "example_04.png"));
OcrPdfCreatorProperties properties = new OcrPdfCreatorProperties();
properties.SetImageLayerName("image");
properties.SetTextLayerName("text");
OcrPdfCreator ocrPdfCreator = new OcrPdfCreator(tesseractReader, properties);
PdfDocument doc = ocrPdfCreator.CreatePdf(files, GetPdfWriter(pdfPath));
NUnit.Framework.Assert.IsNotNull(doc);
int numOfPages = doc.GetNumberOfPages();
NUnit.Framework.Assert.AreEqual(numOfPages, files.Count);
IList <PdfLayer> layers = doc.GetCatalog().GetOCProperties(true).GetLayers();
NUnit.Framework.Assert.AreEqual(numOfPages * 2, layers.Count);
NUnit.Framework.Assert.AreEqual("image", layers[2].GetPdfObject().Get(PdfName.Name).ToString());
NUnit.Framework.Assert.AreEqual("text", layers[3].GetPdfObject().Get(PdfName.Name).ToString());
doc.Close();
// Text layer should contain all text
// Image layer shouldn't contain any text
String expectedOutput = "619121";
NUnit.Framework.Assert.AreEqual(expectedOutput, GetTextFromPdfLayer(pdfPath, "text", 3));
NUnit.Framework.Assert.AreEqual("", GetTextFromPdfLayer(pdfPath, "image", 3));
}
public virtual void TestFontColorInMultiPagePdf()
{
String testName = "testFontColorInMultiPagePdf";
String path = TEST_IMAGES_DIRECTORY + "multîpage.tiff";
String pdfPath = GetTargetDirectory() + testName + ".pdf";
FileInfo file = new FileInfo(path);
tesseractReader.SetTesseract4OcrEngineProperties(tesseractReader.GetTesseract4OcrEngineProperties().SetPreprocessingImages
(false));
OcrPdfCreatorProperties ocrPdfCreatorProperties = new OcrPdfCreatorProperties();
ocrPdfCreatorProperties.SetTextLayerName("Text1");
Color color = DeviceCmyk.MAGENTA;
ocrPdfCreatorProperties.SetTextColor(color);
OcrPdfCreator ocrPdfCreator = new OcrPdfCreator(tesseractReader, ocrPdfCreatorProperties);
PdfDocument doc = ocrPdfCreator.CreatePdf(JavaCollectionsUtil.SingletonList <FileInfo>(file), GetPdfWriter(
pdfPath));
NUnit.Framework.Assert.IsNotNull(doc);
doc.Close();
PdfDocument pdfDocument = new PdfDocument(new PdfReader(pdfPath));
IntegrationTestHelper.ExtractionStrategy strategy = new IntegrationTestHelper.ExtractionStrategy("Text1");
PdfCanvasProcessor processor = new PdfCanvasProcessor(strategy);
processor.ProcessPageContent(pdfDocument.GetPage(1));
Color fillColor = strategy.GetFillColor();
NUnit.Framework.Assert.AreEqual(fillColor, color);
pdfDocument.Close();
}
/// <summary>
/// Perform OCR with custom ocr engine using provided input image and set
/// of properties and save to the given path.
/// </summary>
public static void CreatePdf(String pdfPath, FileInfo inputFile, OcrPdfCreatorProperties properties)
{
OcrPdfCreator ocrPdfCreator = new OcrPdfCreator(new CustomOcrEngine(), properties);
try {
using (PdfWriter pdfWriter = GetPdfWriter(pdfPath)) {
ocrPdfCreator.CreatePdf(JavaCollectionsUtil.SingletonList <FileInfo>(inputFile), pdfWriter).Close();
}
}
catch (System.IO.IOException e) {
LOGGER.Error(e.Message);
}
}
public virtual void TestTextFromPdfLayersFromMultiPageTiff()
{
String testName = "testTextFromPdfLayersFromMultiPageTiff";
bool preprocess = tesseractReader.GetTesseract4OcrEngineProperties().IsPreprocessingImages();
String path = TEST_IMAGES_DIRECTORY + "multîpage.tiff";
String pdfPath = GetTargetDirectory() + testName + ".pdf";
FileInfo file = new FileInfo(path);
tesseractReader.SetTesseract4OcrEngineProperties(tesseractReader.GetTesseract4OcrEngineProperties().SetPreprocessingImages
(false));
OcrPdfCreatorProperties properties = new OcrPdfCreatorProperties();
properties.SetTextLayerName("Text Layer");
properties.SetImageLayerName("Image Layer");
OcrPdfCreator ocrPdfCreator = new OcrPdfCreator(tesseractReader, properties);
PdfDocument doc = ocrPdfCreator.CreatePdf(JavaCollectionsUtil.SingletonList <FileInfo>(file), GetPdfWriter(
pdfPath));
NUnit.Framework.Assert.IsNotNull(doc);
int numOfPages = doc.GetNumberOfPages();
IList <PdfLayer> layers = doc.GetCatalog().GetOCProperties(true).GetLayers();
NUnit.Framework.Assert.AreEqual(numOfPages * 2, layers.Count);
NUnit.Framework.Assert.AreEqual("Image Layer", layers[2].GetPdfObject().Get(PdfName.Name).ToString());
NUnit.Framework.Assert.AreEqual("Text Layer", layers[3].GetPdfObject().Get(PdfName.Name).ToString());
doc.Close();
// Text layer should contain all text
// Image layer shouldn't contain any text
String expectedOutput = "Multipage\nTIFF\nExample\nPage 5";
NUnit.Framework.Assert.AreEqual(expectedOutput, GetTextFromPdfLayer(pdfPath, "Text Layer", 5));
NUnit.Framework.Assert.AreEqual("", GetTextFromPdfLayer(pdfPath, "Image Layer", 5));
NUnit.Framework.Assert.IsFalse(tesseractReader.GetTesseract4OcrEngineProperties().IsPreprocessingImages());
tesseractReader.SetTesseract4OcrEngineProperties(tesseractReader.GetTesseract4OcrEngineProperties().SetPreprocessingImages
(preprocess));
}
public virtual void TestSpanishPNG()
{
String testName = "compareSpanishPNG";
String filename = "scanned_spa_01";
String expectedText1 = "¿Y SI ENSAYARA COMO ACTUAR?";
String expectedText2 = "¿Y SI ENSAYARA ACTUAR?";
String resultPdfPath = GetTargetDirectory() + filename + "_" + testName + "_" + testFileTypeName + ".pdf";
IList <String> languages = JavaUtil.ArraysAsList("spa", "spa_old");
Tesseract4OcrEngineProperties properties = tesseractReader.GetTesseract4OcrEngineProperties();
if (isExecutableReaderType)
{
properties.SetPreprocessingImages(false);
}
// locate text by words
properties.SetTextPositioning(TextPositioning.BY_WORDS);
properties.SetLanguages(languages);
tesseractReader.SetTesseract4OcrEngineProperties(properties);
OcrPdfCreatorProperties ocrPdfCreatorProperties = new OcrPdfCreatorProperties();
ocrPdfCreatorProperties.SetTextColor(DeviceCmyk.BLACK);
OcrPdfCreator ocrPdfCreator = new OcrPdfCreator(tesseractReader, ocrPdfCreatorProperties);
using (PdfWriter pdfWriter = GetPdfWriter(resultPdfPath)) {
ocrPdfCreator.CreatePdf(JavaCollectionsUtil.SingletonList <FileInfo>(new FileInfo(TEST_IMAGES_DIRECTORY + filename
+ ".png")), pdfWriter).Close();
}
try {
String result = GetTextFromPdfLayer(resultPdfPath, null, 1).Replace("\n", " ");
NUnit.Framework.Assert.IsTrue(result.Contains(expectedText1) || result.Contains(expectedText2));
}
finally {
NUnit.Framework.Assert.AreEqual(TextPositioning.BY_WORDS, tesseractReader.GetTesseract4OcrEngineProperties
().GetTextPositioning());
}
}
