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);
}
/// <summary>
/// pulls marked from param inputFilePath image
/// </summary>
/// <param name="fileBytes">Byte array of file data</param>
/// <returns>encodes text</returns>
public static string getText(byte[] fileBytes)
{
string text = "", rootPath = HostingEnvironment.ApplicationPhysicalPath;
BitmapToPixConverter b = new BitmapToPixConverter();
using (Stream memStream = new MemoryStream(fileBytes))
using (Bitmap image = (Bitmap)Image.FromStream(memStream))
using (TesseractEngine ocr = new TesseractEngine(rootPath, "eng", EngineMode.TesseractOnly))
{
image.SetResolution(300, 300);
ocr.SetVariable("tessedit_char_whitelist", "\\/|#");
Pix p = b.Convert(image);
p = p.ConvertRGBToGray();
Page page = ocr.Process(p, PageSegMode.Auto);
text = page.GetText();
p.Dispose();
page.Dispose();
}
return text;
}
public string OCRTiff(string path)
{
String PageText; String Result = String.Empty;
try
{
using (TesseractEngine TE = new TesseractEngine(TessDataPath, "eng", EngineMode.Default))
{
using (Pix Image = Pix.LoadFromFile(TiffPath))
{
PageText = TE.Process(Image).GetText();
}
}
String TextFilePath = string.Format("{0}{1}", TiffPath.Substring(0, TiffPath.Length - 4), ".txt");
File.WriteAllText(TextFilePath, PageText);
Result = "Sucess!";
}
catch (Exception Exception)
{
Result = String.Format("Error: {0}", Exception.Message);
}
return Result;
}
private OCREngine()
{
m_ocrEngine = new TesseractEngine(Configuration.Instance.OCR.TesseractDataPath,
"eng", EngineMode.TesseractAndCube);
m_jsonCardsFilePath = Configuration.Instance.JSONCardsFilePath;
m_cards = new List<string>();
}
public void imageOCRTest(FileDialog openImageDialog )
{
DialogResult ImageResult = openImageDialog.ShowDialog();
if (ImageResult == DialogResult.OK)
{
String testImagePath = openImageDialog.FileName;
try
{
using (var tEngine = new TesseractEngine("C:\\Users\\yeghiakoronian\\Documents\\visual studio 2013\\Projects\\NLP Genre Recogition\\NLP Genre Recogition\\tessdata", "eng", EngineMode.Default)) //creating the tesseract OCR engine with English as the language
{
using (var img = Pix.LoadFromFile(testImagePath)) // Load of the image file from the Pix object which is a wrapper for Leptonica PIX structure
{
using (var page = tEngine.Process(img)) //process the specified image
{
String text = page.GetText(); //Gets the image's content as plain text.
MessageBox.Show(text);
getGenreOfSong(text);
// Console.ReadKey();
}
}
}
}
catch (IOException)
{
MessageBox.Show("Woops Cant Open The File", "COMP 6781: NLP", MessageBoxButtons.OK, MessageBoxIcon.Exclamation);
}
}
}
public static string ReadText(string inputImagePath)
{
var srcImage = new Mat(inputImagePath);
List <Bitmap> bitmap;
var bbox = Cv2.SelectROIs("Select Text Boxes. Enter for confirm, Esc for finish", srcImage);
Cv2.DestroyWindow("Select Text Boxes. Enter for confirm, Esc for finish");
if (bbox == null || bbox.Length == 0)
{
bitmap = new List <Bitmap>()
{
OpenCvSharp.Extensions.BitmapConverter.ToBitmap(srcImage)
}
}
;
else
{
bitmap = bbox.Select(x => OpenCvSharp.Extensions.BitmapConverter.ToBitmap(srcImage.SubMat(x))).ToList();
}
using (var ocr = new Tesseract.TesseractEngine(@"../../../../OpenCVHavrylov/Data/tessdata", "eng"))
{
var res = "";
int i = 1;
foreach (var img in bitmap)
{
using (var page = ocr.Process(img))
{
res += $"[Block {i}]: {page.GetText()}";
i++;
}
}
return(res);
}
}
private string BeginRecognize(string pathToImage)
{
try
{
if (!File.Exists(pathToImage))
return "Image not found";
using (var engine = new TesseractEngine(_pathToTestData, Language, EngineMode.Default))
{
using (var img = Pix.LoadFromFile(pathToImage))
{
using (var page = engine.Process(img))
{
var resultText = page.GetText();
if (!String.IsNullOrEmpty(resultText))
return resultText;
}
}
}
}
catch (Exception e)
{
Trace.TraceError(e.ToString());
MessageBox.Show(e.StackTrace);
return null;
}
return null;
}
private static void Main(string[] args)
{
Console.Write("Input image path:");
string input = Console.ReadLine();
Stopwatch stopwatch = new Stopwatch();
stopwatch.Start();
T.TesseractEngine engine = new T.TesseractEngine("tessdata", "tur", T.EngineMode.TesseractAndLstm);
stopwatch.Stop();
Console.WriteLine("Engine creation :" + stopwatch.ElapsedMilliseconds.ToString() + " ms");
stopwatch.Restart();
T.Pix image = T.Pix.LoadFromFile(input);
T.Page page = engine.Process(image);
string text = page.GetText();
stopwatch.Stop();
Console.WriteLine("Process time :" + stopwatch.ElapsedMilliseconds.ToString() + " ms");
Console.Write("Result: " + text);
Console.Read();
}
/// <summary>
/// Constructor
/// </summary>
public CardRecognizer()
{
//Initialize common filter sequence , this sequence generally will be applied
commonSeq = new FiltersSequence();
commonSeq.Add(Grayscale.CommonAlgorithms.BT709);
commonSeq.Add(new BradleyLocalThresholding());
commonSeq.Add(new DifferenceEdgeDetector());
//Load Templates From Resources ,
//Templates will be used for template matching
j = miranda.ui.Properties.Resources.J;
k = miranda.ui.Properties.Resources.K;
q = miranda.ui.Properties.Resources.Q;
clubs = miranda.ui.Properties.Resources.Clubs;
diamonds = miranda.ui.Properties.Resources.Diamonds;
spades = miranda.ui.Properties.Resources.Spades;
hearts = miranda.ui.Properties.Resources.Hearts;
try
{
_engine = new TesseractEngine(@"./tessdata", "rus", EngineMode.Default);
//_engine.SetVariable("tessedit_char_whitelist", "$.,0123456789");
}
catch (Exception ex)
{
Trace.TraceError(ex.ToString());
Ex.Report(ex);
}
}
public WrapperTesseract(string path, string lng)
{
//Path should be same as TESSDATA folder
//This path should always end with a "/" or "\", e.g., TESSDATA_PREFIX="/usr/share/tesseract-ocr/"
////// Emgu Tesseract(Tesseract3.1)///////
// _ocr = new Emgu.CV.OCR.Tesseract(path, lng, Emgu.CV.OCR.Tesseract.OcrEngineMode.OEM_TESSERACT_CUBE_COMBINED);
// _ocr.SetVariable("tessedit_char_whitelist", "abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ");
// _ocr.SetVariable("user_words_suffix", "user-words");
// _ocr.SetVariable("chop_enable", "1");
// _ocr.SetVariable("tessedit_char_blacklist", "¢§+~»~`!@#$%^&*()_+-={}[]|\\:\";\'<>?,./");
////// Emgu Tesseract(Tesseract3.1)///////
Log.WriteLine("Setting Tesseract traindata and language");
////// Tesseract 3.2 ////////
if (lng=="eng")
_engine = new TesseractEngine(@"./tessdata3/", lng, EngineMode.TesseractAndCube);
else
_engine = new TesseractEngine(@"./tessdata3", lng, EngineMode.Default);
Log.WriteLine("Tesseract Version: " + _engine.Version);
//_engine.SetVariable("tessedit_char_whitelist", "abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ");
//_engine.SetVariable("tessedit_char_blacklist", "¢§+~»~`!@#$%^&*()_+-={}[]|\\:\";\'<>?,./");
////// Tesseract 3.2 ////////
}
/// <summary>
/// 文字認識
/// </summary>
private void AnalyzeString()
{
// Bitmapを処理、適度に切り取った方がOCRを扱いやすい
System.Drawing.Rectangle rect = new System.Drawing.Rectangle(20, 90, 450, 100);
SrcImg = SrcImg.Clone(rect, SrcImg.PixelFormat);
// 確認用の画面に画像を出力
this.Dispatcher.Invoke(() => {
this._Image_crop.Source = ConvertBitmapToIS(SrcImg);
});
// 以下OCR
string langPath = @"C:\tessdata";
string lngStr = "eng";
//画像ファイルでテストするならパス指定
//var img = new Bitmap(@"C:\test.jpg");
var img = SrcImg;
using (var tesseract = new Tesseract.TesseractEngine(langPath, lngStr))
{
// OCRの実行
Pix pix = PixConverter.ToPix(img);
Tesseract.Page page = tesseract.Process(pix);
//表示
Console.WriteLine(page.GetText());
Console.ReadLine(); // err?
}
}
public static string ocr(Bitmap CropedPrescription, int x, int y, int width, int height)
{
string text;
// var PrescriptionImage = CropedPrescription;
using (var engine = new TesseractEngine(@"C:\Program Files\Tesseract-OCR\tessdata\", "kor", EngineMode.Default))
{
engine.SetVariable("tessedit_char_whitelist", "0123456789-."); // 숫자와 . - 만 인식하도록 설정
var roi = new Rect(x, y, width, height); // region of interest 좌표를 생성하고
// using (var img = Pix.LoadFromFile(PrescriptionImage))
// {
using (var page = engine.Process(CropedPrescription, roi, PageSegMode.SingleLine))
{
text = page.GetText();
System.IO.File.WriteAllText(@"C:\Program Files\PLOCR\textrecognition.html", text); // 인식한 글자를 html 형식으로 저장한다.
// text = TextProcess.RemoveWhiteSpace(text);
// Console.WriteLine("인식한 문자: \n{0}\n", text);
// Console.Read();
}
// }
}
return text;
}
public override string ExtractContent(int? pageNumber)
{
var dataPath = Config.TessDataPath;
if (!Directory.Exists(dataPath))
{
throw new ArgumentException("Path does not exist or access is denied.", nameof(dataPath));
}
using (var engine = new TesseractEngine(dataPath, "eng", EngineMode.Default))
{
using (var memoryStream = new MemoryStream(Buffer))
{
// have to load Pix via a bitmap since Pix doesn't support loading a stream.
using (var image = new Bitmap(memoryStream))
{
using (var pix = PixConverter.ToPix(image))
{
using (var page = engine.Process(pix))
{
return page.GetText();
}
}
}
}
}
}
/// <summary>
/// Charactor Recognition Proccess
/// </summary>
/// <param name="imgPath"></param>
/// <param name="dataDirPath"></param>
/// <param name="lang"></param>
/// <returns></returns>
private static string Character_recognition_Process(string imgPath, string dataDirPath, string lang)
{
if (!System.IO.File.Exists(imgPath))
{
//Console.Error.WriteLine("画像のパスに画像が見つかりませんでした");
return("画像のパスに画像が見つかりませんでした");
}
//Reading language learning data
string traindedDataPath = System.IO.Path.Combine(dataDirPath, lang + ".traineddata");
if (!System.IO.File.Exists(traindedDataPath))
{
//Console.Error.WriteLine(lang + ".traineddataがみつかりませんでした");
return(lang + ".traineddataがみつかりませんでした");
}
// Character recognition
using (TesseractEngine tesseract = new Tesseract.TesseractEngine(dataDirPath, lang))
{
// Import an image file
var img = new System.Drawing.Bitmap(imgPath);
// Specifying a character
tesseract.SetVariable("tessedit_char_whitelist", "1234567890ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz.,");
// PERFORM OCR
Tesseract.Page page = tesseract.Process(img);
//Console.WriteLine(page.GetText());
return(page.GetText());
}
}
public static string GetOCRText(string engpath, Image imgdata)
{
string rst = "";
try
{
Directory.SetCurrentDirectory(System.IO.Path.GetDirectoryName(Environment.GetCommandLineArgs()[0]));
using (var engine = new TesseractEngine(engpath, "eng", EngineMode.Default))
{
using (Bitmap image = new Bitmap(imgdata))
{
using (var pix = PixConverter.ToPix(image))
{
using (var page = engine.Process(pix))
{
rst = page.GetText().Trim();
}
}
}
}
}
catch (System.Exception ex)
{
ScratchModel.WriteLogFile("----Error----", "------ GetOCRText1 -------", "");
}
return rst;
}
private void btnOCR_Click(object sender, EventArgs e)
{
try
{
string testImagePath = this.txtFile.Text.Trim();
using (var engine = new TesseractEngine(@"./tessdata", "eng", EngineMode.Default))
{
using (var img = Pix.LoadFromFile(testImagePath))
{
System.Diagnostics.Debug.WriteLine("Process image");
var i = 1;
using (var page = engine.Process(img))
{
var text = page.GetText();
System.Diagnostics.Debug.WriteLine("Text: {0}", text);
System.Diagnostics.Debug.WriteLine("Mean confidence: {0}", page.GetMeanConfidence());
using (var iter = page.GetIterator())
{
iter.Begin();
do
{
if (i % 2 == 0)
{
System.Diagnostics.Debug.WriteLine("Line {0}", i);
do
{
System.Diagnostics.Debug.WriteLine("Word Iteration");
if (iter.IsAtBeginningOf(PageIteratorLevel.Block))
{
System.Diagnostics.Debug.WriteLine("New block");
}
if (iter.IsAtBeginningOf(PageIteratorLevel.Para))
{
System.Diagnostics.Debug.WriteLine("New paragraph");
}
if (iter.IsAtBeginningOf(PageIteratorLevel.TextLine))
{
System.Diagnostics.Debug.WriteLine("New line");
}
System.Diagnostics.Debug.WriteLine("word: " + iter.GetText(PageIteratorLevel.Word));
} while (iter.Next(PageIteratorLevel.TextLine, PageIteratorLevel.Word));
}
i++;
} while (iter.Next(PageIteratorLevel.Para, PageIteratorLevel.TextLine));
}
}
}
}
}
catch (Exception ex)
{
Console.WriteLine("Unexpected Error: " + ex.Message);
Console.WriteLine("Details: ");
Console.WriteLine(ex.ToString());
}
}
// 辨識驗證碼
private string parseCaptchaStr(Image image)
{
Tesseract.TesseractEngine ocr = new Tesseract.TesseractEngine(@"tessdata\", "yzufont");
ocr.SetVariable("tessedit_char_whitelist", "0123456789ABCDEFGHIJKLMNOPQRSTUVWXYZ");
Page tmpPage = ocr.Process(new Bitmap(image), pageSegMode: ocr.DefaultPageSegMode);
return(Regex.Replace(tmpPage.GetText(), @"\s+", String.Empty));
}
internal Page(TesseractEngine engine, Pix image, string imageName, Rect regionOfInterest, PageSegMode pageSegmentMode)
{
Engine = engine;
Image = image;
ImageName = imageName;
RegionOfInterest = regionOfInterest;
PageSegmentMode = pageSegmentMode;
}
/// <summary>
/// Creates OCR engine
/// </summary>
/// <returns></returns>
private Tesseract.TesseractEngine CreateEngine()
{
Tesseract.TesseractEngine tesseractEngine = new Tesseract.TesseractEngine("tessdata", GetLanguageNamesString(RecognitionLanguage1, RecognitionLanguage2), EngineMode.ToEngineMode());
if (SearchForDigitsOnly)
{
tesseractEngine.SetVariable("tessedit_char_whitelist", "0123456789");
}
return(tesseractEngine);
}
/// <summary>
/// Processes the specified image path.
/// </summary>
/// <param name="bitmap">Bitmap.</param>
/// <returns></returns>
public string Process(Bitmap bitmap)
{
using (Tesseract.TesseractEngine engine = CreateEngine())
{
engine.DefaultPageSegMode = (PageSegMode)DefaultSegmentationMode;
using (Page page = engine.Process(bitmap))
return(page.GetText());
}
}
public void TestInitialize()
{
// OCRを行うオブジェクトの生成
// 言語データの場所と言語名を引数で指定する
var path = testPath + @"\tessdata";
// 英語なら"eng" 「○○.traineddata」の○○の部分
tesseractJpn = new TesseractEngine(path, "jpn");
tesseractEng = new TesseractEngine(path, "eng");
}
//// Do training for all existing trained Data
public SVM(string TrainedDataInputFile)
{
_engine = new TesseractEngine(@"./tessdata3", "eng", EngineMode.TesseractAndCube);
_engine.SetVariable("tessedit_char_whitelist", "abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ");
_engine.SetVariable("tessedit_char_blacklist", "¢§+~»~`!@#$%^&*()_+-={}[]|\\:\";\'<>?,./");
string[] TrainedData = Directory.GetFiles(TrainedDataInputFile, "*.png");
double[][] inputs = new double[TrainedData.Length][]; ///
double[] InputArray = new double[784];
int[] Outputs = new int[TrainedData.Length];
for (int i = 0; i < TrainedData.Length; i++)
{
string filename = Path.GetFileNameWithoutExtension(TrainedData[i]);
Bitmap TrainingImage = new Bitmap(TrainedData[i]);
string[] split = filename.Split('.');
for (int j = 0; j < 28; j++)
{
for (int k = 0; k < 28; k++)
{
if ((!TrainingImage.GetPixel(j, k).Name.Equals("ffffffff")))
InputArray[j * 28 + k] = 1;
else
InputArray[j * 28 + k] = 0;
}
}
inputs[i] = InputArray;
Outputs[i] = Convert.ToInt32(split[0]);
InputArray = new double[784];
}
IKernel kernel;
kernel = new Polynomial(2, 0);
ksvm = new MulticlassSupportVectorMachine(784, kernel, 2);
MulticlassSupportVectorLearning ml = new MulticlassSupportVectorLearning(ksvm, inputs, Outputs);
double complexity = 1; ///// set these three parameters Carefuly later
double epsilon = 0.001;
double tolerance = 0.2;
ml.Algorithm = (svm, classInputs, classOutputs, i, j) =>
{
var smo = new SequentialMinimalOptimization(svm, classInputs, classOutputs);
smo.Complexity = complexity; /// Cost parameter for SVM
smo.Epsilon = epsilon;
smo.Tolerance = tolerance;
return smo;
};
// Train the machines. It should take a while.
double error = ml.Run();
}
private string OCR(Bitmap b)
{
string res = "";
using (var engine = new TesseractEngine(@"tessdata", "eng", EngineMode.Default))
{
engine.SetVariable("tessedit_char_whitelist", "1234567890ABCDEFGHIJKLMNOPQRSTUVWXYZ");
engine.SetVariable("tessedit_unrej_any_wd", true);
using (var page = engine.Process(b, PageSegMode.SingleLine))
res = page.GetText();
}
return res;
}
public static async Task init()
{
try {
string exe_dir = (new FileInfo(System.Reflection.Assembly.GetEntryAssembly().Location)).Directory.FullName;
string tessdata_dir = Path.Combine(exe_dir, "tessdata");
engine = new TesseractEngine(tessdata_dir, "eng", EngineMode.TesseractAndCube);
} catch (Exception e) {
Trace.TraceError(e.ToString());
Console.WriteLine("Unexpected Error: " + e.Message);
Console.WriteLine("Details: ");
Console.WriteLine(e.ToString());
}
}
private void LoadButton_OnClick(object sender, RoutedEventArgs e)
{
var fileDialog = new OpenFileDialog();
var showDialog = fileDialog.ShowDialog();
if (showDialog != null && showDialog.Value)
{
//@"D:\HandwriteExpressionRecognition\HandwriteExpressionRecognition\HandwriteExpressionRecognition.Desktop\TestData\123456789.bmp"
var tessEngine = new Tesseract.TesseractEngine("tessdata", "eng", EngineMode.Default);
var page = tessEngine.Process(
Pix.LoadFromFile(fileDialog.FileName),
PageSegMode.Auto);
TextBlock.Text = page.GetText();
}
}
private static string RecognizeTextFromImage(Image image)
{
string recognizedText;
using (var ocrEngine = new TesseractEngine(@"./tessdata", "eng"))
using (Bitmap bitmap = new Bitmap(image))
using (Bitmap monochromeBitmap = ImageUtils.Convert24BitToMonochrome(bitmap))
using (var page = ocrEngine.Process(monochromeBitmap))
{
recognizedText = page.GetText();
Trace.TraceInformation("Text recognized with mean confidence: {0:N3}", page.GetMeanConfidence());
}
return recognizedText;
}
public static string AnalyzeFileHOCR(string p, string fileName)
{
var data = System.IO.Path.Combine(p, "tessdata");
using (var engine = new TesseractEngine(data, "fra", EngineMode.Default))
{
using (var img = Pix.LoadFromFile(fileName))
{
using (var page = engine.Process(img))
{
return page.GetHOCRText(0);
}
}
}
}
public vhConsole(IConfig cfg, String uHash)
{
vhConsole.uHash = uHash;
config = cfg;
//var info = MyInfo().Result;
//vhConsole.uHash = (string)info["uhash"];
if (!Directory.Exists(cfg.tessdata))
{
cfg.logger.Log("Cannot find tessdata path: {0}", Path.GetFullPath(cfg.tessdata));
//throw new Exception();
}
engine = new Tesseract.TesseractEngine(cfg.tessdata, "eng");
}
static void Main(string[] args)
{
var image = new Bitmap(@"C:\\Users\\E566281\\Documents\\2007_AEB_E-pdf-1.pdf");
var ocr = new Tesseract.TesseractEngine("C:\\Users\\E566281\\Documents\\2007_AEB_E-pdf-1.pdf", "eng");
var result = ocr.Process(image, null);
foreach (var s in result.GetHOCRText(1, false))
{
Console.WriteLine(s);
}
//var result = ocr ( image, Rectangle.Empty);
//foreach (tessnet2.Word word in result)
//{
// Console.writeline(word.text);
//}
}
/// <summary>
/// Gets all the numbers seperately from image.
/// </summary>
/// <param name="imagePath">The image path.</param>
/// <returns></returns>
public static List<int> GetAllNumbersFromImage(string imagePath)
{
var numbersList = new List<int>();
if (string.IsNullOrWhiteSpace(imagePath) == false && System.IO.File.Exists(imagePath) == true)
{
try
{
using (var engine = new TesseractEngine(@"./tessdata", "eng", EngineMode.Default))
{
using (var img = Pix.LoadFromFile(imagePath))
{
using (var page = engine.Process(img))
{
if (string.IsNullOrWhiteSpace(page.GetText()) == false)
{
page.GetText()
.Split(' ')
.ToList()
.ForEach(text =>
{
if (string.IsNullOrWhiteSpace(text) == false)
{
var numberValue = 0;
text = regex.Replace(text, string.Empty); //remove non alpha numeric characters
text = text.ToLowerInvariant().Replace('i', '1').Replace('o', '0'); //to fix wrong interpretation
if (int.TryParse(text, out numberValue) == false)
{
Console.WriteLine("Unable to process the file : " + imagePath);
}
else
{
numbersList.Add(numberValue);
}
}
});
}
}
}
}
}
catch (Exception ex)
{
Console.WriteLine("Unexpected Error: " + ex.Message);
}
}
return numbersList;
}
public static string imageToText(string tiffPath)
{
string text = "";
int stop = 0;
try
{
using (var engine = new TesseractEngine(@"./tessdata", "eng", EngineMode.Default))
{
using (var img = Pix.LoadFromFile(tiffPath))
{
var i = 1;
using (var page = engine.Process(img))
{
text = page.GetText();
Debug.WriteLine("Mean confidence: {0}", page.GetMeanConfidence());
using (var iter = page.GetIterator())
{
iter.Begin();
do
{
if (i % 2 == 0)
{
do
{
text = text + iter.GetText(PageIteratorLevel.Word);
} while (iter.Next(PageIteratorLevel.TextLine, PageIteratorLevel.Word));
}
i++;
} while (iter.Next(PageIteratorLevel.Para, PageIteratorLevel.TextLine));
iter.Dispose();
}
}
}
}
}
catch (Exception e)
{
Debug.WriteLine("Unexpected Error: " + e.Message);
Debug.WriteLine("Details: ");
Debug.WriteLine(e.ToString());
}
return text;
}
private void BTN_Save_Click_1(object sender, EventArgs e)
{
string filename = path + @"\img_ocr" + count.ToString();
camera.Capture(filename);
Bitmap img = (Bitmap)Bitmap.FromFile(filename + ".jpg");
img.RotateFlip(RotateFlipType.Rotate180FlipY);
Tesseract.TesseractEngine tesseract = new Tesseract.TesseractEngine("./dataset", "eng", Tesseract.EngineMode.TesseractOnly);
var page = tesseract.Process(img);
listBox1.Items.Add(page.GetText());
img.Dispose();
File.Delete(filename + ".jpg");
count++;
}
static void Main(string[] args)
{
while (true)
{
videoSource = new VideoCaptureDevice(new FilterInfoCollection(FilterCategory.VideoInputDevice)[0].MonikerString);
videoSource.NewFrame += CaptureFrame;
videoSource.Start();
Thread.Sleep(1000);
videoSource.Stop();
var engine = new TesseractEngine(@"tessdata", "letsgodigital", EngineMode.Default);
var image = new ElitechStc1000Image("test3.jpg", engine);
var temp = image.GetTemperature();
Console.WriteLine(temp);
try
{
var sensorEvent = new SensorEvent
{
Temperature = temp
};
Console.WriteLine("{0} > Sending temperature: {1}", DateTime.Now, temp);
var sink = new AzuresStreamAnalytics()
.WithConnectionString(ConfigurationManager.AppSettings["EventHubSendConnection"]);
sink.Send(sensorEvent);
}
catch (Exception exception)
{
Console.ForegroundColor = ConsoleColor.Red;
Console.WriteLine("{0} > Exception: {1}", DateTime.Now, exception.Message);
Console.ResetColor();
}
Thread.Sleep(2000);
}
}
public string GetOCRText(string engpath, Image imgdata)
{
string rst = "";
using (var engine = new TesseractEngine(engpath, "eng", EngineMode.Default))
{
using (Bitmap image = new Bitmap(imgdata))
{
using (var pix = PixConverter.ToPix(image))
{
using (var page = engine.Process(pix))
{
//meanConfidenceLabel.InnerText = String.Format("{0:P}", page.GetMeanConfidence());
rst = page.GetText().Trim();
}
}
}
}
return rst;
}
public void Test()
{
var testImagePath = "../../phototest.tif";
try
{
using (var engine = new TesseractEngine(@"../../tessdata", "eng", EngineMode.Default))
{
using (var img = Pix.LoadFromFile(testImagePath))
{
var i = 1;
using (var page = engine.Process(img))
{
var text = page.GetText();
using (var iter = page.GetIterator())
{
iter.Begin();
do
{
if (i % 2 == 0)
{
do
{
Console.WriteLine("word: " + iter.GetText(PageIteratorLevel.Word));
} while (iter.Next(PageIteratorLevel.TextLine, PageIteratorLevel.Word));
}
i++;
} while (iter.Next(PageIteratorLevel.Para, PageIteratorLevel.TextLine));
}
}
}
}
}
catch (Exception e)
{
Console.WriteLine("Unexpected Error: " + e.Message);
Console.WriteLine("Details: ");
Console.WriteLine(e.ToString());
}
}
public string GetText(Bitmap image)
{
using (var tesseract = new TesseractEngine(_dataPath, EngineLang, EngineMode.Default))
{
tesseract.SetVariable("tessedit_write_images", true);
using (var pix = PixConverter.ToPix(image))
{
using (var page = tesseract.Process(pix))
{
var text = RemoveSpecialCharacters(page.GetText());
#if DEBUG
System.Diagnostics.Debug.WriteLine("Confibialidade: " + page.GetMeanConfidence());
System.Diagnostics.Debug.WriteLine("Placa: " + text);
#endif
return text;
}
}
}
}
public string DownloadAndProcessImage(Uri uri, Rect regionPercentage)
{
var engine = new TesseractEngine(_tessdataPath, "hun");
byte[] imageData;
try
{
var client = new WebClient();
imageData = client.DownloadData(uri);
StatisticsCollector.GetInstance().IncrementOutgoingWebRequestCount();
_log.Debug("Image downloaded from `" + uri + "`. Size: " + imageData.Length + " Byte(s).");
}
catch (Exception e)
{
_log.Error("Failed to download image from `" + uri + "`. Error: e");
return "";
}
var bitmap = new Bitmap(new MemoryStream(imageData));
_log.Debug("Image size: " + bitmap.Size);
// Calculate image region
var region = new Tesseract.Rect(
(int) (regionPercentage.Left*bitmap.Width),
(int) (regionPercentage.Top*bitmap.Height),
(int) (regionPercentage.Width*bitmap.Width),
(int) (regionPercentage.Height*bitmap.Height));
_log.Debug("OCR region: " + region);
Page page = engine.Process(bitmap, region, PageSegMode.Auto);
string text = page.GetText();
_log.Debug("Processed text length: " + text.Length);
StatisticsCollector.GetInstance().IncrementOcredPageCount();
return text;
}
// 특정 좌표 지역을 받아서 그 부분만 판독하는 함수
public static string hocr(Bitmap source, int x, int y, int width, int height)
{
string htext;
// var PrescriptionImage = CropedPrescription;
using (var engine = new TesseractEngine(@"C:\Program Files\Tesseract-OCR\tessdata\", "kor", EngineMode.Default))
{
// using (var img = Pix.LoadFromFile(PrescriptionImage)
// {
var roi = new Rect(x, y, width, height); // region of interest 좌표를 생성하고
using (var page = engine.Process(source, roi, PageSegMode.Auto))
{
htext = page.GetHOCRText(3);
System.IO.File.WriteAllText(@"C:\Program Files\PLOCR\textrecognition.html", htext); // 인식한 글자를 html 형식으로 저장한다.
// Console.WriteLine(htext);
// Console.Read();
}
// }
return htext;
}
}
/// <summary>
/// Método responsavel por realizar a quebra do captcha
/// </summary>
/// <param name="image">Byte Array da imagem captcha</param>
/// <param name="config">Objeto contendo informações relevantes para solucionar o captcha</param>
/// <returns>String com a solução do captcha</returns>
public string Break(byte[] image, CaptchaConfig config)
{
var tempfilename = Path.GetTempFileName();
File.WriteAllBytes(tempfilename, image);
var stbResult = new StringBuilder();
using (var engine = new TesseractEngine(ConfigManager.GetInstance().CAMINHO_ARQUIVO_IDIOMA, IDIOMA_PORTUGUES, EngineMode.Default))
using (var img = Pix.LoadFromFile(tempfilename))
{
var i = 1;
using (var page = engine.Process(img, null))
{
stbResult.AppendLine(page.GetText());
if (config.NumeroDePalavras > 1)
{
using (var iter = page.GetIterator())
{
iter.Begin();
do
{
if (i % 2 == 0)
{
do
{
stbResult.AppendLine(iter.GetText(PageIteratorLevel.Word));
} while (iter.Next(PageIteratorLevel.TextLine, PageIteratorLevel.Word));
}
i++;
} while (iter.Next(PageIteratorLevel.Para, PageIteratorLevel.TextLine));
}
}
}
}
return stbResult.ToString().Replace("\r", string.Empty).Replace("\n", string.Empty); ;
}
private void OcrRecognize(string path)
{
_tesseractEngine = new TesseractEngine(tessData, Lang, EngineMode.Default);
_words = new List<Line>();
var img = Pix.LoadFromFile(path);
var page = _tesseractEngine.Process(img);
var text = page.GetText();
using (var iter = page.GetIterator())
{
iter.Begin();
Rect rect = new Rect();
do
{
iter.TryGetBoundingBox(PageIteratorLevel.TextLine, out rect);
_words.Add(new Line(rect.X1, rect.Y1, rect.Width, rect.Height, iter.GetText(PageIteratorLevel.TextLine),path));
//System.Diagnostics.Debug.WriteLine("X:{0}, Y:{1}", rect.X1, rect.Y1);
} while (iter.Next(PageIteratorLevel.Block, PageIteratorLevel.TextLine));
}
img.Dispose();
page.Dispose();
_tesseractEngine.Dispose();
}
public static string AnalyzeFile(string key, string p, string fileName)
{
var data = System.IO.Path.Combine(p, "tessdata");
using (var engine = new TesseractEngine(data, "fra", EngineMode.Default))
{
string bitmapPath = System.IO.Path.Combine(p, Guid.NewGuid().ToString() + System.IO.Path.GetExtension(fileName));
System.IO.MemoryStream ms = new System.IO.MemoryStream(Utility.Security.AES.DecryptFile(key, fileName));
System.Drawing.Bitmap bitmap = new System.Drawing.Bitmap(ms);
bitmap = Treatment.SetContrast(bitmap, 20);
bitmap.Save(bitmapPath);
ms.Close();
using (var img = Pix.LoadFromFile(bitmapPath))
{
using (var page = engine.Process(img))
{
System.IO.File.Delete(bitmapPath);
return page.GetText().Trim();
}
}
}
}
