private string Ocr(Pix target, out double confidence)
{
try
{
var text = "";
using (var engine = new TesseractEngine(@"./tessdata", "eng", EngineMode.Default))
{
using (var page = engine.Process(target))
{
text = page.GetText();
confidence = page.GetMeanConfidence();
}
target.Dispose();
}
return(text);
}
catch (Exception e)
{
Trace.TraceError(e.ToString());
Debug.WriteLine("Unexpected Error: " + e.Message);
Debug.WriteLine("Details: ");
Debug.WriteLine(e.ToString());
}
confidence = 0;
return("");
}
public static OcrResult PopulateMatrixFromImage(Image image)
{
Pix pix = LoadBitmapToPix(image as Bitmap);
List <string> Cells = GetCellsFromPix(pix);
var root = (int)Math.Floor(Math.Sqrt(Cells.Count));
var result = new OcrResult(root);
for (int cell = 0, row = 0; cell < Cells.Count; cell++)
{
if (cell % root == 0 && cell != 0)
{
row++;
}
if (row == root)
{
break;
}
result.SetCell(row, (int)Math.Floor((decimal)cell % root), Cells[cell]);
}
image.Dispose();
pix.Dispose();
return(result);
}
public string OCRimage(Image ImageToUse, int Zoomlevel, out double TimeTaken, out float Confidence)
{
DateTime Starttime = DateTime.Now;
//load in image
LoadImage(ImageToUse);
//post process the image
ImageToOCR = AccordImageProcessing.AccordImageProc.ImageProcessing(ImageToOCR, Zoomlevel);
//Convert to Tesseract format
byte[] ImgByte = ToByteArray(ImageToOCR, System.Drawing.Imaging.ImageFormat.Bmp);
Pix img = Pix.LoadFromMemory(ImgByte);
// OCR it
Page page = TesseractOCRCore.Process(img);
//get test
string text = page.GetText();
//Get confidence
Confidence = page.GetMeanConfidence();
//Get Time
DateTime EndTime = DateTime.Now;
TimeTaken = (EndTime - Starttime).TotalSeconds;
page.Dispose();
img.Dispose();
return(text);
}
public TesseractOCRResult OCRimage(Image ImageToUse, int Zoomlevel)
{
DateTime Starttime = DateTime.Now;
//load in image
LoadImage(ImageToUse);
//post process the image
ImageToOCR = AccordImageProcessing.AccordImageProc.ImageProcessing(ImageToOCR, Zoomlevel);
//Convert to Tesseract format
Pix img = PixConverter.ToPix(ImageToOCR);
// OCR it
Page page = TesseractOCRCore.Process(img);
//get test
string text = page.GetText();
//Get confidence
float Confidence = page.GetMeanConfidence();
//Get Time
DateTime EndTime = DateTime.Now;
double TimeTaken = (EndTime - Starttime).TotalSeconds;
page.Dispose();
img.Dispose();
return(new TesseractOCRResult(text, TimeTaken, Confidence));
}
/// <summary>
/// Destroys
/// <see cref="Tesseract.Pix"/>
/// object.
/// </summary>
/// <param name="pix">
///
/// <see cref="Tesseract.Pix"/>
/// object to be destroyed
/// </param>
internal static void DestroyPix(Pix pix)
{
if (pix != null)
{
pix.Dispose();
}
}
public static IRecognitionResult ReadFile(Pix img, TesseractEngine engine)
{
Stopwatch timer = new Stopwatch();
timer.Start();
try
{
using (Page page = engine.Process(img))
{
string text = page.GetText();
float meanConf = page.GetMeanConfidence();
return(new SimpleRecognitionResult(text, meanConf));
}
}
catch (Exception e)
{
Console.WriteLine(e);
return(new ErrorResult(e));
}
finally
{
timer.Stop();
Console.WriteLine("Processing time: " + timer.Elapsed);
bool imageIsNotDisposed = img != null && !img.IsDisposed;
Console.WriteLine("imageIsNotDisposed " + imageIsNotDisposed);
if (imageIsNotDisposed)
{
img.Dispose();
}
}
}
public List <string> ocrDetectString(string filename)
{
List <string> result = new List <string>();
if (ocr == null)
{
return(result);
}
Pix pix = Pix.LoadFromFile(filename);
if (pix == null)
{
return(result);
}
Page page = ocr.Process(pix);
Pix processed = page.GetThresholdedImage();
processed.Save("./adtemp/process.png");
string text = page.GetText();
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));
}
Console.WriteLine(text);
result.Add(text);
page.Dispose();
pix.Dispose();
return(result);
}
string DoOCR(Pix pix)
{
gOcr.DefaultPageSegMode = PageSegMode.SingleBlock;
Page pg = gOcr.Process(pix);
String buf = pg.GetText();
pix.Dispose();
pg.Dispose();
return(buf.Replace("ی", "ي").Replace("ه", "ە").Replace("\n", Environment.NewLine));
}
public static Pix ToPix(Bitmap img)
{
int pixDepth = GetPixDepth(img.PixelFormat);
Pix pix = Pix.Create(img.Width, img.Height, pixDepth);
pix.XRes = (int)Math.Round(img.HorizontalResolution);
pix.YRes = (int)Math.Round(img.VerticalResolution);
BitmapData bitmapData = null;
try
{
if ((img.PixelFormat & PixelFormat.Indexed) == PixelFormat.Indexed)
{
CopyColormap(img, pix);
}
bitmapData = img.LockBits(new Rectangle(0, 0, img.Width, img.Height), ImageLockMode.ReadOnly, img.PixelFormat);
PixData data = pix.GetData();
if (bitmapData.PixelFormat == PixelFormat.Format32bppArgb)
{
TransferDataFormat32bppArgb(bitmapData, data);
}
else if (bitmapData.PixelFormat == PixelFormat.Format32bppRgb)
{
TransferDataFormat32bppRgb(bitmapData, data);
}
else if (bitmapData.PixelFormat == PixelFormat.Format24bppRgb)
{
TransferDataFormat24bppRgb(bitmapData, data);
}
else if (bitmapData.PixelFormat == PixelFormat.Format8bppIndexed)
{
TransferDataFormat8bppIndexed(bitmapData, data);
}
else if (bitmapData.PixelFormat == PixelFormat.Format1bppIndexed)
{
TransferDataFormat1bppIndexed(bitmapData, data);
}
return(pix);
}
catch (Exception)
{
pix.Dispose();
throw;
}
finally
{
if (bitmapData != null)
{
img.UnlockBits(bitmapData);
}
}
}
private List <string> run_tessract()
{
List <string> re = new List <string>();
string dataPath = "./tessdata/";
string language = 1 != 1 ? "eng" : "jpn";
string inputFile = "./_tmp.bmp";
OcrEngineMode oem = OcrEngineMode.DEFAULT;
PageSegmentationMode psm = PageSegmentationMode.AUTO_OSD;
TessBaseAPI tessBaseAPI = new TessBaseAPI();
// Initialize tesseract-ocr
if (!tessBaseAPI.Init(dataPath, language, oem))
{
throw new Exception("Could not initialize tesseract.");
}
// Set the Page Segmentation mode
tessBaseAPI.SetPageSegMode(psm);
// Set the input image
Pix pix = tessBaseAPI.SetImage(inputFile);
tessBaseAPI.SetVariable("number", "1234567890");
// Recognize image
tessBaseAPI.Recognize();
ResultIterator resultIterator = tessBaseAPI.GetIterator();
// extract text from result iterator
StringBuilder stringBuilder = new StringBuilder();
PageIteratorLevel pageIteratorLevel = PageIteratorLevel.RIL_PARA;
do
{
string str = resultIterator.GetUTF8Text(pageIteratorLevel);
if (str != null)
{
str = Regex.Replace(str, @"\n", "\r\n");
re.Add(str);
}
} while (resultIterator.Next(pageIteratorLevel));
tessBaseAPI.Dispose();
pix.Dispose();
return(re);
}
/// <summary>
/// Runs a basic OCR on an image.
/// </summary>
/// <param name="pixImage">The image.</param>
/// <param name="dataPath">The directory holding the tesseract data.</param>
/// <param name="language">The language to use during the OCR process.</param>
/// <param name="disposeImage">If true, the image will be disposed after the OCR process is done.</param>
/// <returns></returns>
public static string GetText(this Pix pixImage,
string dataPath = "./tessdata",
string language = "eng",
bool disposeImage = true)
{
using var engine = new TesseractEngine(dataPath, language, EngineMode.Default);
using var process = engine.Process(pixImage);
var text = process.GetText();
if (disposeImage)
{
pixImage.Dispose();
}
return(text);
}
/// <summary>
/// Ensures data folder exists, and run an OCR.
/// </summary>
/// <param name="pixImage">The image.</param>
/// <param name="dataPath">The directory holding the tesseract data.</param>
/// <param name="language">The language to use during the OCR process.</param>
/// <param name="disposeImage">If true, the image will be disposed after the OCR process is done.</param>
/// <returns></returns>
public static async Task <string> GetTextAndEnsureData(this Pix pixImage,
string dataPath = "./tessdata",
string language = "eng",
bool disposeImage = true)
{
var tessdataDownloader = new TessdataDownloader();
await tessdataDownloader.EnsureDataFolder(language, dataFolder : dataPath);
using var engine = new TesseractEngine(dataPath, language, EngineMode.Default);
using var process = engine.Process(pixImage);
var text = process.GetText();
if (disposeImage)
{
pixImage.Dispose();
}
return(text);
}
//Get word location
public string OCRimagewithLocation(Image ImageToUse, int Zoomlevel)
{
//load in image
LoadImage(ImageToUse);
//post process the image
ImageToOCR = AccordImageProcessing.AccordImageProc.ImageProcessing(ImageToOCR, Zoomlevel);
//Convert to Tesseract format
Pix img = PixConverter.ToPix(ImageToOCR);
// OCR it
Page page = TesseractOCRCore.Process(img);
//get test
string text = page.GetHOCRText(1);
img.Dispose();
page.Dispose();
return(text);
//return "";
}
static void example4()
{
string dataPath = "./tessdata/";
string language = "eng";
string inputFile = "./input.png";
OcrEngineMode oem = OcrEngineMode.DEFAULT;
PageSegmentationMode psm = PageSegmentationMode.AUTO_OSD;
TessBaseAPI tessBaseAPI = new TessBaseAPI();
// Initialize tesseract-ocr
if (!tessBaseAPI.Init(dataPath, language, oem))
{
throw new Exception("Could not initialize tesseract.");
}
// Set the Page Segmentation mode
tessBaseAPI.SetPageSegMode(psm);
// Set the input image
Pix pix = tessBaseAPI.SetImage(inputFile);
// Recognize image
tessBaseAPI.Recognize();
//ensure input name is set
tessBaseAPI.SetInputName(inputFile);
var fileInfo = new System.IO.FileInfo(inputFile);
string tessDataPath = tessBaseAPI.GetDatapath();
string outputName = fileInfo.FullName.Replace(fileInfo.Extension, string.Empty); //input name.pdf
// call pdf renderer and export pdf
using (var pdfRenderer = new PdfRenderer(outputName, tessDataPath, false))
{
pdfRenderer.BeginDocument("tesseract.net searchable Pdf generation");
pdfRenderer.AddImage(tessBaseAPI);
pdfRenderer.EndDocument();
}
tessBaseAPI.Dispose();
pix.Dispose();
}
//Get word location
public string OCRimagewithLocation(Image ImageToUse, int Zoomlevel)
{
//load in image
LoadImage(ImageToUse);
//post process the image
ImageToOCR = AccordImageProcessing.AccordImageProc.ImageProcessing(ImageToOCR, Zoomlevel);
//Convert to Tesseract format
byte[] ImgByte = ToByteArray(ImageToOCR, System.Drawing.Imaging.ImageFormat.Bmp);
Pix img = Pix.LoadFromMemory(ImgByte);
// OCR it
Page page = TesseractOCRCore.Process(img);
//get test
string text = page.GetHOCRText(1);
img.Dispose();
page.Dispose();
return(text);
}
private void FillPagesData(ATAPY.Document.Data.Core.Document document, string pathToImage, string language)
{
var extension = Path.GetExtension(pathToImage);
using (var engine = new TesseractEngine(ENGINE_DATAPATH, language))
{
if (IsSinglePageImage(extension))
{
Pix pageData = null;
try
{
pageData = Pix.LoadFromFile(pathToImage);
FillDocumentPage(document, language, engine, pageData);
}
finally
{
pageData?.Dispose();
}
}
else if (IsMultiPageImage(extension))
{
PixArray pixes = null;
try
{
pixes = PixArray.LoadMultiPageTiffFromFile(pathToImage);
foreach (Pix pageData in pixes)
{
FillDocumentPage(document, language, engine, pageData);
}
}
finally
{
pixes?.Dispose();
}
}
else
{
throw new FormatException("Please specify path to the image file");
}
}
}
static void example3()
{
string dataPath = "./tessdata/";
//string language = "eng";
string language = "chi_sim";
string inputFile = "./input.png";
OcrEngineMode oem = OcrEngineMode.DEFAULT;
PageSegmentationMode psm = PageSegmentationMode.AUTO_OSD;
TessBaseAPI tessBaseAPI = new TessBaseAPI();
// Initialize tesseract-ocr
if (!tessBaseAPI.Init(dataPath, language, oem))
{
throw new Exception("Could not initialize tesseract.");
}
// Set the Page Segmentation mode
tessBaseAPI.SetPageSegMode(psm);
// Set the input image
Pix pix = tessBaseAPI.SetImage(inputFile);
// Recognize image
tessBaseAPI.Recognize();
ResultIterator resultIterator = tessBaseAPI.GetIterator();
// extract text from result iterator
StringBuilder stringBuilder = new StringBuilder();
PageIteratorLevel pageIteratorLevel = PageIteratorLevel.RIL_PARA;
do
{
stringBuilder.Append(resultIterator.GetUTF8Text(pageIteratorLevel));
} while (resultIterator.Next(pageIteratorLevel));
tessBaseAPI.Dispose();
pix.Dispose();
}
static void Main(string[] args)
{
new DirectoryInfo(@"C:\Users\ACER\Desktop\Tool")
.GetFiles("*.jpg", SearchOption.AllDirectories)
.ToList()
.ForEach(x => {
Pix pixs = Pix.LoadFromFile(x.FullName);
pixs = pixs.PixBackgroundNormSimple();
pixs = pixs.PixConvertRGBToGray();
pixs = pixs.PixFindSkewAndDeskew();
pixs = pixs.PixTophat();
pixs = pixs.PixInvert();
pixs = pixs.PixGammaRTC();
pixs = pixs.PixThresholdToBinary();
/*
* pixs = Pix.Create(LeptonicaNativeApi.Native.pixBackgroundNormSimple(pixs.Reference, IntPtr.Zero, IntPtr.Zero));
* pixs = Pix.Create(LeptonicaNativeApi.Native.pixConvertRGBToGray(pixs.Reference, 0.5f, 0.3f, 0.2f));
*
* float angle, confidence;
* pixs = Pix.Create(LeptonicaNativeApi.Native.pixFindSkewAndDeskew(pixs.Reference, 2, out angle, out confidence));
*
* if(confidence > 2 && confidence < 3) {
* angle *= (float) Math.PI / 180.0f;
* pixs = Pix.Create(LeptonicaNativeApi.Native.pixRotate(pixs.Reference, angle, RotationMethod.Shear, RotationFill.White, pixs.Width, pixs.Height));
* }
*
* pixs = Pix.Create(LeptonicaNativeApi.Native.pixTophat(pixs.Reference, 17, 17, L_TOPHAT.BLACK));
* LeptonicaNativeApi.Native.pixInvert(pixs.Reference, pixs.Reference);
* LeptonicaNativeApi.Native.pixGammaTRC(pixs.Reference, pixs.Reference, 1.0f, 170, 245);
*
* pixs = Pix.Create(LeptonicaNativeApi.Native.pixThresholdToBinary(pixs.Reference, 35));*/
pixs.Save(x.ToNewExtension(".pdf").FullName, ImageSaveFormat.Lpdf);
pixs.Dispose();
});
Console.WriteLine("Entre key to exit...");
Console.ReadLine();
}
//public static void clearFaceFlag() { faceFlag = 0; }
public string RecognizeText(int id)
{
DBService dbs = new DBService();
Document doc = dbs.FindDocumentById(id);
Image <Gray, Byte> img = scale(doc);
//var image = PixConverter.ToPix(img.ToBitmap()))
Tesseract.Native.DllImports.TesseractDirectory = System.Web.HttpContext.Current.Server.MapPath("~/Tesseract/bin/Debug/DLLS/");
TessBaseAPI tessBaseAPI = new TessBaseAPI();
System.Diagnostics.Debug.WriteLine("The current version is {0}", tessBaseAPI.GetVersion());
string dataPath = System.Web.HttpContext.Current.Server.MapPath("~/tessdata/");
string language = "eng";
string inputFile = doc.Path;
OcrEngineMode oem = OcrEngineMode.DEFAULT;
//OcrEngineMode oem = OcrEngineMode.DEFAULT;
PageSegmentationMode psm = PageSegmentationMode.AUTO_OSD;
// Initialize tesseract-ocr
if (!tessBaseAPI.Init(dataPath, language, oem))
{
throw new Exception("Could not initialize tesseract.");
}
// Set the Page Segmentation mode
tessBaseAPI.SetPageSegMode(psm);
// Set the input image
Pix pix = tessBaseAPI.SetImage(inputFile);
// Recognize image
tessBaseAPI.Recognize();
ResultIterator resultIterator = tessBaseAPI.GetIterator();
// extract text from result iterator
StringBuilder stringBuilder = new StringBuilder();
int top, bottom, left, right, i = 0;
List <OCRText> forJson = new List <OCRText>();
PageIteratorLevel pageIteratorLevel = PageIteratorLevel.RIL_TEXTLINE;
do
{
string textContent = resultIterator.GetUTF8Text(pageIteratorLevel);
resultIterator.BoundingBox(pageIteratorLevel, out left, out top, out right, out bottom);
string coordsString = "" + left + "," + top + "," + right + "," + bottom;
forJson.Add(new OCRText()
{
Coords = coordsString, Text = textContent
});
} while (resultIterator.Next(pageIteratorLevel));
tessBaseAPI.Dispose();
pix.Dispose();
var textForReturn = JsonConvert.SerializeObject(forJson);
dbs.UpdateDocument(textForReturn, id);
if (HasFace(img) == true)
{
FaceFlag = 1;
}
else
{
FaceFlag = 0;
}
return(textForReturn);
}
public new void KeyDown(KeyboardHookEventArgs e)
{
// Use Singleton design to store location information
PartLocation currentLocation = PartLocation._instance;
// Print Screen key was pressed
if (e.Key == Keys.PrintScreen)
{
// Grab image from screen and convert to black and white using a threshold
PrintScreenThreshold();
// Process text from image
try
{
string dataPath = @"./tessdata";
string language = "eng";
string inputFile = "rewards.jpg";
OcrEngineMode oem = OcrEngineMode.DEFAULT;
PageSegmentationMode psm = PageSegmentationMode.SINGLE_LINE;
TessBaseAPI tessBaseAPI = new TessBaseAPI();
// Initialize tesseract-ocr
if (!tessBaseAPI.Init(dataPath, language, oem))
{
throw new Exception("Could not initialize tesseract");
}
// Set the Page Segmentation mode
tessBaseAPI.SetPageSegMode(psm);
// Warframe Relics are displayed using A-Z, a-z, and &
tessBaseAPI.SetVariable("tessedit_char_whitelist", "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz&");
// Set input file
Pix pix = tessBaseAPI.SetImage(inputFile);
TB_Part1.Text = GetText(tessBaseAPI, currentLocation.PartLoc1, 580);
TB_Part2.Text = GetText(tessBaseAPI, currentLocation.PartLoc2, 580);
if (currentLocation.NumPeople >= 3)
{
TB_Part3.Text = GetText(tessBaseAPI, currentLocation.PartLoc3, 580);
}
if (currentLocation.NumPeople == 4)
{
TB_Part4.Text = GetText(tessBaseAPI, currentLocation.PartLoc4, 580);
}
tessBaseAPI.Dispose();
pix.Dispose();
// Grab latest order list for parts
// TODO: Try multiple threads! If that's too fast, cache prices for a duration
using (var client = new WebClient())
{
GetPriceJson(client, TB_Part1, TB_Plat1);
GetPriceJson(client, TB_Part2, TB_Plat2);
if (currentLocation.NumPeople >= 3)
{
GetPriceJson(client, TB_Part3, TB_Plat3);
}
if (currentLocation.NumPeople == 4)
{
GetPriceJson(client, TB_Part4, TB_Plat4);
}
// These get cached
GetDucatsJson(client, TB_Part1, TB_Ducats1);
GetDucatsJson(client, TB_Part2, TB_Ducats2);
if (currentLocation.NumPeople >= 3)
{
GetDucatsJson(client, TB_Part3, TB_Ducats3);
}
if (currentLocation.NumPeople == 4)
{
GetDucatsJson(client, TB_Part4, TB_Ducats4);
}
}
List <PartWorth> sortPartWorth = new List <PartWorth>();
sortPartWorth.Add(new PartWorth(TB_Part1, TB_Ducats1));
sortPartWorth.Add(new PartWorth(TB_Part2, TB_Ducats2));
if (currentLocation.NumPeople >= 3)
{
sortPartWorth.Add(new PartWorth(TB_Part3, TB_Ducats3));
}
if (currentLocation.NumPeople == 4)
{
sortPartWorth.Add(new PartWorth(TB_Part4, TB_Ducats4));
}
// Sort by Plat, then Ducats
sortPartWorth = sortPartWorth.OrderByDescending(o => Int32.Parse(o.platinum.Tag.ToString())).ThenByDescending(o => Int32.Parse(o.ducats.Tag.ToString())).ToList();
// If max Plat is low, sort by the reverse
if (Int32.Parse(sortPartWorth[0].platinum.Tag.ToString()) < 15)
{
sortPartWorth = sortPartWorth.OrderByDescending(o => Int32.Parse(o.ducats.Tag.ToString())).ThenByDescending(o => Int32.Parse(o.platinum.Tag.ToString())).ToList();
}
// Show best option
TB_Pick.Text = sortPartWorth[0].platinum.Text;
// Show current amount of Plat (and Ducats) made in the current session
if (Int32.Parse(sortPartWorth[0].platinum.Tag.ToString()) != -1)
{
//GlobalVar.PLAT += Int32.Parse(platDuc[0].plat.Tag.ToString());
LB_Plat.Tag = Int32.Parse(LB_Plat.Tag.ToString()) + Int32.Parse(sortPartWorth[0].platinum.Tag.ToString());
}
if (Int32.Parse(sortPartWorth[0].ducats.Tag.ToString()) != -1)
{
//GlobalVar.DUCATS += Int32.Parse(platDuc[0].ducats.Tag.ToString());
LB_Ducs.Tag = Int32.Parse(LB_Ducs.Tag.ToString()) + Int32.Parse(sortPartWorth[0].ducats.Tag.ToString());
}
LB_Plat.Text = LB_Plat.Tag.ToString() + " p";
LB_Ducs.Text = LB_Ducs.Tag.ToString() + " d";
}
catch (Exception ex)
{
LogError("KeyDown: " + ex.Message);
}
}
else if (e.Key == Keys.NumPad2)
{
Debug.WriteLine("Switched to 2 people");
ClearBox();
VisibleRow(false, LB_Part3, TB_Part3, TB_Plat3, TB_Ducats3);
VisibleRow(false, LB_Part4, TB_Part4, TB_Plat4, TB_Ducats4);
currentLocation.SetPeople2();
}
else if (e.Key == Keys.NumPad3)
{
Debug.WriteLine("Switched to 3 people");
ClearBox();
VisibleRow(true, LB_Part3, TB_Part3, TB_Plat3, TB_Ducats3);
VisibleRow(false, LB_Part4, TB_Part4, TB_Plat4, TB_Ducats4);
currentLocation.SetPeople3();
}
else if (e.Key == Keys.NumPad4)
{
Debug.WriteLine("Switched to 4 people");
ClearBox();
VisibleRow(true, LB_Part3, TB_Part3, TB_Plat3, TB_Ducats3);
VisibleRow(true, LB_Part4, TB_Part4, TB_Plat4, TB_Ducats4);
currentLocation.SetPeople4();
}
else if (e.Key == Keys.Pause)
{
Environment.Exit(0);
}
}
protected override string DoScan(string filename, LoggingSection log)
{
PdfDocument pdfDocument = PdfReader.Open(filename);
StringBuilder stringBuilder = new StringBuilder();
for (int pageIndex = 0; pageIndex < pdfDocument.PageCount; pageIndex++)
{
log.Verbose($"Scanning page {pageIndex + 1} of {pdfDocument.PageCount}");
PdfPage pdfPage = pdfDocument.Pages[pageIndex];
//Extract text from text elements
stringBuilder.Append($"{ExtractTextFromPdfPage(pdfPage)}{Environment.NewLine}");
//Extract text from image elements with Tesseract OCR - awesome! :)
PdfDictionary resources = pdfPage.Elements.GetDictionary("/Resources");
if (resources != null)
{
PdfDictionary xObjects = resources.Elements.GetDictionary("/XObject");
if (xObjects != null)
{
ICollection <PdfItem> items = xObjects.Elements.Values;
foreach (PdfItem item in items)
{
PdfReference reference = item as PdfReference;
if (reference != null)
{
PdfDictionary xObject = reference.Value as PdfDictionary;
if (xObject != null && xObject.Elements.GetString("/Subtype") == "/Image")
{
Bitmap bitmap = PdfImageToBitmap(xObject);
if (bitmap == null)
{
log.Error("Could not extract bitmap from PDF image element. Seems like the PDF image filter type is not supported. Skipping element!");
continue;
}
log.Debug("Rotating image");
bitmap.RotateFlip(RotateFlipType.Rotate90FlipNone);
log.Debug("Upscaling image 2x");
BitmapUtils.Scale(ref bitmap, 2);
log.Debug("Grayscaling image");
BitmapUtils.GrayscaleWithLockBits(bitmap);
log.Debug("Denoising image");
BitmapUtils.DenoiseWithLockBits(bitmap);
log.Debug("Applying OCR on image");
Pix pix = PixConverter.ToPix(bitmap);
TesseractEngine tesseractEngine = Services.OCRProvider.AwaitResource();
Page tesseractPage = tesseractEngine.Process(pix);
try
{
string text = tesseractPage.GetText();
log.Debug($"Text is {text.Length} characters long");
if (!string.IsNullOrWhiteSpace(text) && text != "\n")
{
stringBuilder.Append(text.Replace("\n", " "));
}
}
catch (InvalidOperationException e)
{
log.Error($"OCR failed on Page {pageIndex} of file {filename}:\n{e.StackTrace}");
}
Services.OCRProvider.Feed(tesseractEngine);
pix.Dispose();
}
}
}
}
}
stringBuilder.Append("\n");
}
log.Debug("Trimming text");
string documentText = stringBuilder.ToString();
documentText = documentText.Trim();
while (documentText.Contains(" "))
{
documentText = documentText.Replace(" ", " ");
}
while (documentText.Contains("\n\n"))
{
documentText = documentText.Replace("\n\n", "\n");
}
return(stringBuilder.ToString());
}
