public void Print(ResultIterator iter)
{
logger.Log("Is beginning of block: {0}", iter.IsAtBeginningOf(PageIteratorLevel.Block));
logger.Log("Is beginning of para: {0}", iter.IsAtBeginningOf(PageIteratorLevel.Para));
logger.Log("Is beginning of text line: {0}", iter.IsAtBeginningOf(PageIteratorLevel.TextLine));
logger.Log("Is beginning of word: {0}", iter.IsAtBeginningOf(PageIteratorLevel.Word));
logger.Log("Is beginning of symbol: {0}", iter.IsAtBeginningOf(PageIteratorLevel.Symbol));
logger.Log("Block text: \"{0}\"", iter.GetText(PageIteratorLevel.Block));
logger.Log("Para text: \"{0}\"", iter.GetText(PageIteratorLevel.Para));
logger.Log("TextLine text: \"{0}\"", iter.GetText(PageIteratorLevel.TextLine));
logger.Log("Word text: \"{0}\"", iter.GetText(PageIteratorLevel.Word));
logger.Log("Symbol text: \"{0}\"", iter.GetText(PageIteratorLevel.Symbol));
}
public List <Rectangle> GetTextRects(Bitmap currentImage)
{
string tessPath = System.IO.Path.GetDirectoryName(System.Reflection.Assembly.GetExecutingAssembly().Location) + "\\tessdata";
TesseractEngine tess = new TesseractEngine(tessPath, "eng");
Page newPage = tess.Process(currentImage, PageSegMode.AutoOsd);
ResultIterator iterator = newPage.GetIterator();
string totalText = newPage.GetText();
List <Rectangle> currentRects = new List <Rectangle>();
for (int i = 0; i < totalText.Length; i++)
{
Rect foundRect = new Rect();
string symbolText = iterator.GetText(PageIteratorLevel.Symbol);
bool hasText = symbolText != "" && symbolText != null;
bool onlyLetters = false;
if (hasText)
{
onlyLetters = symbolText.ToCharArray().All(s => char.IsLetter(s));
}
bool gotBoundingBox = iterator.TryGetBoundingBox(PageIteratorLevel.Symbol, out foundRect);
if (hasText && onlyLetters && gotBoundingBox)
{
currentRects.Add(new Rectangle(foundRect.X1, foundRect.Y1, foundRect.X2 - foundRect.X1, foundRect.Y2 - foundRect.Y1));
}
iterator.Next(PageIteratorLevel.Symbol);
}
tess.Dispose();
iterator.Dispose();
return(currentRects);
}
private static IEnumerable <RecognizedTextChunk> recognizeWords(PdfPage page, TesseractEngine engine,
int resolution, string tempFileName)
{
// Save PDF page as high-resolution image
PdfDrawOptions options = PdfDrawOptions.Create();
options.BackgroundColor = new PdfRgbColor(255, 255, 255);
options.HorizontalResolution = resolution;
options.VerticalResolution = resolution;
page.Save(tempFileName, options);
using (var img = Pix.LoadFromFile(tempFileName))
{
using (var recognizedPage = engine.Process(img))
{
using (ResultIterator iter = recognizedPage.GetIterator())
{
const PageIteratorLevel Level = PageIteratorLevel.Word;
iter.Begin();
do
{
if (iter.TryGetBoundingBox(Level, out Rect bounds))
{
string text = iter.GetText(Level);
float confidence = iter.GetConfidence(Level);
yield return(new RecognizedTextChunk(text, bounds, confidence));
}
} while (iter.Next(Level));
}
}
}
}
private static TextBlockItem ExtractParaphraph(ResultIterator iter)
{
var text = new StringBuilder();
do
{
do
{
text.Append(iter.GetText(PageIteratorLevel.Word));
text.Append(" ");
if (iter.IsAtFinalOf(PageIteratorLevel.TextLine, PageIteratorLevel.Word))
{
text.Append(Environment.NewLine);
}
}while (iter.Next(PageIteratorLevel.TextLine, PageIteratorLevel.Word));
if (iter.IsAtFinalOf(PageIteratorLevel.Para, PageIteratorLevel.TextLine))
{
text.Append(Environment.NewLine);
}
}while (iter.Next(PageIteratorLevel.Para, PageIteratorLevel.TextLine));
return(new TextBlockItem
{
Text = text.ToString()
});
}
public static string RecognizeBlocks(byte[] imageBytes)
{
StringBuilder sb = new StringBuilder();
try
{
using (var engine = new TesseractEngine(@"./tessdata", "eng", EngineMode.Default))
{
using (Pix pix = Pix.LoadFromMemory(imageBytes))
{
using (var page = engine.Process(pix))
{
sb.AppendLine("Text (iterator):");
using (ResultIterator iter = page.GetIterator())
{
iter.Begin();
do
{
do
{
do
{
do
{
if (iter.IsAtBeginningOf(PageIteratorLevel.Block))
{
sb.AppendLine("<BLOCK>");
}
sb.Append(iter.GetText(PageIteratorLevel.Word));
sb.Append(" ");
if (iter.IsAtFinalOf(PageIteratorLevel.TextLine, PageIteratorLevel.Word))
{
sb.AppendLine();
}
} while (iter.Next(PageIteratorLevel.TextLine, PageIteratorLevel.Word));
if (iter.IsAtFinalOf(PageIteratorLevel.Para, PageIteratorLevel.TextLine))
{
sb.AppendLine();
}
} 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());
}
return(sb.ToString());
}
public static string OCR(Bitmap imagem, string linguagem)
{
string texto = "";
using (TesseractEngine engine = new TesseractEngine(@"C:\GitHub\operacao-politica-supervisionada\OPS\temp\", linguagem, EngineMode.Default)) {
engine.SetVariable("tessedit_char_whitelist", "1234567890ABCDEFGHIJKLMNOPQRSTUVWXYZ");
engine.SetVariable("tessedit_unrej_any_wd", true);
engine.SetVariable("applybox_learn_chars_and_char_frags_mode", true);
engine.SetVariable("save_blob_choices", true);
string sobreposto = "";
int ultimo = 12;
using (Page page = engine.Process(imagem, PageSegMode.SingleLine)) {
using (ResultIterator ri = page.GetIterator()) {
do
{
string word = ri.GetText(PageIteratorLevel.Symbol);
Tesseract.Rect bb;
if (ri.TryGetBoundingBox(PageIteratorLevel.Symbol, out bb))
{
if ((bb.Width > 13) && (bb.Height > 15) && (word.Trim() != ""))
{
while (bb.X1 > ultimo + 14)
{
texto += Resolver(sobreposto);
sobreposto = "";
ultimo += 28;
}
//System.Web.HttpContext.Current.Response.Write(word + ": " + bb.X1 + "<br />\n");
if ((word != "Q") || (bb.Height <= 30))
{
sobreposto += word;
}
else
{
sobreposto += "O";
}
}
}
} while((ri.Next(PageIteratorLevel.Symbol)));
if (texto.Length < 6)
{
texto += Resolver(sobreposto);
while (texto.Length < 6)
{
texto += LetraAleatoria();
}
}
}
}
}
return(texto);
}
private void GetPageData(TesseractEngine engine, Pix pageData, string language, ATAPY.Document.Data.Core.Page page)
{
ResultIterator resultIterator = null;
try
{
using (var tessPage = engine.Process(pageData))
{
tessPage.Recognize();
resultIterator = tessPage.GetIterator();
resultIterator.Begin();
do
{
var text = resultIterator.GetText(PageIteratorLevel.Word);
if (TextIsValid(text) && resultIterator.TryGetBoundingBox(PageIteratorLevel.Word, out var rect))
{
var rectW = GetRect(rect);
var area = new TextArea(rectW, text, page);
page.TextAreas.Add(area);
var chars = new System.Windows.Rect[text.Length];
int charIter = 0;
do
{
if (resultIterator.TryGetBoundingBox(PageIteratorLevel.Symbol, out var sRect))
{
chars[charIter] = GetRect(sRect);
}
charIter++;
} while (resultIterator.Next(PageIteratorLevel.Word, PageIteratorLevel.Symbol));
area.SetCharProperties(chars);
}
} while (resultIterator.Next(PageIteratorLevel.Word));
}
}
finally
{
resultIterator?.Dispose();
}
//return page;
}
private static void IterateBlocks(Page page)
{
Console.WriteLine("Text (iterator):");
using (ResultIterator 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));
}
}
public static void Main(string[] args)
{
var testImagePath = "./phototest.tif";
if (args.Length > 0)
{
testImagePath = args[0];
}
try {
var logger = new FormattedConsoleLogger();
var resultPrinter = new ResultPrinter(logger);
using (var engine = new TesseractEngine(@"./tessdata", "eng", EngineMode.Default)) {
using (var img = Pix.LoadFromFile(testImagePath)) {
using (logger.Begin("Process image")) {
var i = 1;
using (var page = engine.Process(img)) {
var text = page.GetText();
logger.Log("Text: {0}", text);
logger.Log("Mean confidence: {0}", page.GetMeanConfidence());
using (var iter = page.GetIterator()) {
iter.Begin();
do
{
if (i % 2 == 0)
{
using (logger.Begin("Line {0}", i)) {
do
{
using (logger.Begin("Word Iteration")) {
if (iter.IsAtBeginningOf(PageIteratorLevel.Block))
{
logger.Log("New block");
}
if (iter.IsAtBeginningOf(PageIteratorLevel.Para))
{
logger.Log("New paragraph");
}
if (iter.IsAtBeginningOf(PageIteratorLevel.TextLine))
{
logger.Log("New line");
}
logger.Log("word: " + iter.GetText(PageIteratorLevel.Word));
ResultIterator testiter = iter.Clone();
logger.Log("from clone: " + testiter.GetText(PageIteratorLevel.Word));
}
} while (iter.Next(PageIteratorLevel.TextLine, PageIteratorLevel.Word));
}
}
i++;
} while (iter.Next(PageIteratorLevel.Para, PageIteratorLevel.TextLine));
}
}
}
}
}
} 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);
}
public static string Box(Bitmap imagem, string linguagem, string correto, int pagina)
{
string texto = "";
using (TesseractEngine engine = new TesseractEngine(@"C:\GitHub\operacao-politica-supervisionada\OPS\temp\", linguagem)) {
engine.SetVariable("tessedit_char_whitelist", "1234567890ABCDEFGHIJKLMNOPQRSTUVWXYZ");
engine.SetVariable("tessedit_unrej_any_wd", true);
engine.SetVariable("applybox_learn_chars_and_char_frags_mode", true);
engine.SetVariable("save_blob_choices", true);
string sobreposto = "";
int ultimo = 12;
using (Page page = engine.Process(imagem, PageSegMode.SingleLine)) {
Tesseract.Rect bb;
int x1 = 14, y1 = 0, x2 = 0, y2 = 0;
int pos = 0;
int miny1 = 50;
using (ResultIterator ri = page.GetIterator()) {
do
{
string word = ri.GetText(PageIteratorLevel.Symbol);
if (ri.TryGetBoundingBox(PageIteratorLevel.Symbol, out bb))
{
if ((bb.Width > 13) && (bb.Height > 15) && (word.Trim() != ""))
{
while (bb.X1 > ultimo + 14)
{
x2 = Math.Max(x1 + 15, x2);
texto += correto[pos] + " " + x1 + " " + Math.Min(10, y1) + " " + x2 + " " + Math.Max(40, y2) + " " + pagina + "\n";
pos++;
sobreposto = "";
ultimo += 28;
x1 = Math.Max(x1 + 28, x2);
}
miny1 = Math.Min(miny1, bb.Y1);
if (sobreposto != "")
{
x1 = Math.Min(x1, bb.X1);
y1 = Math.Min(y1, bb.Y1);
x2 = Math.Max(x2, bb.X2);
y2 = Math.Max(y2, bb.Y2);
}
else
{
x1 = Math.Max(x2 - 5, bb.X1);
y1 = bb.Y1;
x2 = bb.X2;
y2 = bb.Y2;
}
//System.Web.HttpContext.Current.Response.Write(word + ": " + bb.X1 + "<br />\n");
if ((word != "Q") || (bb.Height <= 30))
{
sobreposto += word;
}
else
{
sobreposto += "O";
}
}
}
} while((ri.Next(PageIteratorLevel.Symbol)));
int limite = imagem.Width - 6;
if (pos < 6)
{
texto += correto[pos] + " " + x1 + " " + Math.Min(10, y1) + " " + x2 + " " + Math.Max(40, y2) + " " + pagina + "\n";
while (pos < 5)
{
pos++;
x1 = x2;
x2 = x1 + (limite - x1) / (6 - pos);
texto += correto[pos] + " " + x1 + " " + Math.Min(10, y1) + " " + x2 + " " + Math.Max(40, y2) + " " + pagina + "\n";
}
}
}
if (miny1 > 40)
{
texto = "";
}
}
}
return(texto);
}
/*
* int LookingTextMarker(RecognazeRule Rule, Page page, out ResultIterator BestLineIter, out int word)
* {
* word = -1;
* BestLineIter = null;
* int BestDistance = 10000;
*
* ResultIterator LineIter = page.GetIterator();
* string[] Words = Rule.TextMarker.Split(new char[] {' '}, StringSplitOptions.RemoveEmptyEntries);
* int NumberOfWords = Words.Length;
* LineIter.Begin();
* do
* {
* int CurrentWordNumber = -1;
* int CurrentBestDistance = 10000;
* string Line = LineIter.GetText(PageIteratorLevel.TextLine);
* if(Line == null)
* continue;
* string[] WordsOfLine = Line.Split(new char[] {' '}, StringSplitOptions.None);
* if(WordsOfLine.Length < NumberOfWords)
* continue;
*
* for(int shift = 0; shift <= WordsOfLine.Length - NumberOfWords; shift++)
* {
* int PassDistance = 0;
* for(int i = 0; i < NumberOfWords; i++)
* {
* PassDistance += FuzzyStringComparer.GetDistanceLevenshtein(WordsOfLine[shift + i],
* Words[i],
* StringComparison.CurrentCultureIgnoreCase);
* }
* if(PassDistance < CurrentBestDistance)
* {
* CurrentBestDistance = PassDistance;
* CurrentWordNumber = shift + 1;
* }
* }
* if(CurrentBestDistance < BestDistance)
* {
* AddToLog ("new best");
* AddToLog (LineIter.GetText(PageIteratorLevel.Word));
* word = CurrentWordNumber;
* if(BestLineIter != null)
* BestLineIter.Dispose();
* BestLineIter = LineIter.Clone();
* AddToLog (BestLineIter.GetText(PageIteratorLevel.TextLine));
* BestDistance = CurrentBestDistance;
* }
* } while( LineIter.Next(PageIteratorLevel.TextLine));
* LineIter.Dispose();
* return BestDistance;
* } */
int GetTextPosition(string Text, Page page, out int PosX, out int PosY, out double AngleRad, RecognazeRule[] AfterMarkerRules)
{
int BestDistance = 10000;
PosX = -1;
PosY = -1;
AngleRad = 0;
logger.Debug("Marker zone text:{0}", page.GetText());
ResultIterator LineIter = page.GetIterator();
string[] Words = Text.Split(new char[] { ' ' }, StringSplitOptions.RemoveEmptyEntries);
int NumberOfWords = Words.Length;
LineIter.Begin();
do
{
int CurrentWordNumber = -1;
int CurrentAfterWord = 0;
int CurrentBestDistance = 10000;
string Line = LineIter.GetText(PageIteratorLevel.TextLine);
if (Line == null || Line == "")
{
continue;
}
Line = Line.Trim();
string[] WordsOfLine = Line.Split(new char[] { ' ' }, StringSplitOptions.RemoveEmptyEntries);
if (WordsOfLine.Length == 0)
{
continue;
}
for (int shift = 0; shift < WordsOfLine.Length; shift++)
{
for (int i = 1; i <= NumberOfWords && i <= WordsOfLine.Length - shift; i++)
{
string passString = String.Join(" ", WordsOfLine, shift, i);
int PassDistance = FuzzyStringComparer.GetDistanceLevenshtein(passString,
Text,
StringComparison.CurrentCultureIgnoreCase);
if (PassDistance < CurrentBestDistance)
{
CurrentBestDistance = PassDistance;
CurrentWordNumber = shift;
CurrentAfterWord = shift + i;
}
}
}
if (CurrentBestDistance < BestDistance)
{
logger.Debug("new best");
logger.Debug(LineIter.GetText(PageIteratorLevel.TextLine).Trim());
//Заполняем поля данными после маркера.
foreach (RecognazeRule rule in AfterMarkerRules)
{
if (rule.NextAfterTextMarker && WordsOfLine.Length > CurrentAfterWord + rule.ShiftWordsCount)
{
rule.AfterTextMarkerValue = WordsOfLine[CurrentAfterWord + rule.ShiftWordsCount];
}
}
BestDistance = CurrentBestDistance;
for (int i = 0; i < CurrentWordNumber; i++)
{
LineIter.Next(PageIteratorLevel.Word);
}
Rect Box;
LineIter.TryGetBoundingBox(PageIteratorLevel.Word, out Box);
PosX = Box.X1;
PosY = Box.Y1;
logger.Debug("Position X1:{0} Y1:{1} X2:{2} Y2:{3}", Box.X1, Box.Y1, Box.X2, Box.Y2);
LineIter.TryGetBaseline(PageIteratorLevel.Word, out Box);
logger.Debug("BaseLine X1:{0} Y1:{1} X2:{2} Y2:{3}", Box.X1, Box.Y1, Box.X2, Box.Y2);
AngleRad = Math.Atan2(Box.Y2 - Box.Y1, Box.X2 - Box.X1); //угл наклона базовой линии.
double AngleGrad = AngleRad * (180 / Math.PI);
logger.Debug("Angle rad:{0} grad:{1}", AngleRad, AngleGrad);
//Получаем уровень распознования полей в маркере.
int iterAlreadyShifted = CurrentWordNumber - CurrentAfterWord;
bool stopIteration = false;
foreach (RecognazeRule rule in AfterMarkerRules.Where(x => x.NextAfterTextMarker).OrderBy(x => x.ShiftWordsCount))
{
while (iterAlreadyShifted < rule.ShiftWordsCount)
{
if (LineIter.IsAtFinalOf(PageIteratorLevel.TextLine, PageIteratorLevel.Word))
{
stopIteration = true;
break;
}
LineIter.Next(PageIteratorLevel.Word);
iterAlreadyShifted++;
}
if (stopIteration)
{
break;
}
rule.AfterTextMarkerConfidence = LineIter.GetConfidence(PageIteratorLevel.Word);
logger.Debug("Cлово {0} со сдвигом {1} имеет точность {2}.", LineIter.GetText(PageIteratorLevel.Word), rule.ShiftWordsCount, rule.AfterTextMarkerConfidence);
}
}
}while(LineIter.Next(PageIteratorLevel.TextLine));
LineIter.Dispose();
return(BestDistance);
}