/// <summary>
/// Invokes Tesseract OCR Recognize on given image
/// Stores the resulting data in the Text,Confidence and ScanTime data members
/// </summary>
/// <param name="image"> Source Image to be OCR-ed </param>
/// <returns> Text from Image </returns>
public string Scan(Image<Gray, byte> image)
{
_tesseract.Recognize(image);
_text = _tesseract.GetText();
_chars = _tesseract.GetCharactors();
return _text;
}
private static string OcrInitalizer()
{
_ocr = new Tesseract("tessdata", "eng", Tesseract.OcrEngineMode.OEM_TESSERACT_CUBE_COMBINED);
_ocr.Init(@"C:\Users\Nino\Documents\PbpHackthone\dotNet_OCR_server\Emgu\emgucv-windows-x64 2.4.0.1717\bin", "eng", Tesseract.OcrEngineMode.OEM_DEFAULT);
string result = string.Empty;
try
{
var image = new Image <Bgr, byte>("IMAG0253_cropped.jpg");
using (Image <Gray, byte> gray = image.Convert <Gray, Byte>())
{
gray.Save(@"C:\Users\Nino\Desktop\IMAG0253_cropped.jpg");
Stopwatch watch = new Stopwatch();
watch.Start();
_ocr.Recognize(gray);
Tesseract.Charactor[] charactors = _ocr.GetCharactors();
//String text = String.Concat( Array.ConvertAll(charactors, delegate(Tesseract.Charactor t) { return t.Text; }) );
result = _ocr.GetText();
watch.Stop();
var total = watch.ElapsedMilliseconds;
}
}
catch (Exception exception)
{
}
return(result);
}
private void Form1_Load(object sender, EventArgs e)
{
if (openImageFileDialog.ShowDialog() == System.Windows.Forms.DialogResult.OK)
{
Bgr drawColor = new Bgr(Color.Blue);
try
{
Image <Bgr, Byte> image = new Image <Bgr, byte>(openImageFileDialog.FileName);
original.Image = image.ToBitmap();
original.SizeMode = PictureBoxSizeMode.Zoom;
using (Image <Gray, byte> gray = image.Convert <Gray, Byte>())
{
_ocr.Recognize(gray);
Tesseract.Charactor[] charactors = _ocr.GetCharactors();
foreach (Tesseract.Charactor c in charactors)
{
image.Draw(c.Region, drawColor, 1);
}
processed.Image = image.ToBitmap();
processed.SizeMode = PictureBoxSizeMode.Zoom;
//String text = String.Concat( Array.ConvertAll(charactors, delegate(Tesseract.Charactor t) { return t.Text; }) );
String text = _ocr.GetText();
ocrTextBox.Text = text;
}
}
catch (Exception exception)
{
MessageBox.Show(exception.Message);
}
}
}
public void TestOCREngBlankPage()
{
Version version = Tesseract.Version;
int i = version.Major;
using (Tesseract ocr = GetTesseract())
using (Image <Gray, Byte> img = new Image <Gray, byte>(1024, 960))
{
ocr.Recognize(img);
Tesseract.Charactor[] results = ocr.GetCharactors();
EmguAssert.IsTrue(results.Length == 0);
}
}
public override Rectangle[] find(Image <Bgr, Byte> image)
{
ocr.Recognize(image.Convert <Gray, Byte>());
Tesseract.Charactor[] charactors = ocr.GetCharactors();
result = ocr.GetText();
List <Rectangle> rects = new List <Rectangle>();
foreach (Tesseract.Charactor c in charactors)
{
rects.Add(c.Region);
}
return(rects.ToArray());
}
/// <summary>
/// Recognition of MaterialID with tesseract
/// - Preprocess image (auto-cropping)
/// - Check if count of characters is ok.
/// </summary>
/// <param name="labelBox">box where label is found</param>
/// <param name="img"></param>
/// <param name="bestPoint"></param>
/// <returns></returns>
private string recognizeMaterialID(MCvBox2D labelBox, Image <Bgr, Byte> img, PointW bestPoint)
{
Image <Gray, Byte> tmp = img.Copy(labelBox).Convert <Gray, Byte>();
tmp = tmp.Resize(120, 80, INTER.CV_INTER_LINEAR);
// Auto cropping (only materialID is left)
tmp = preprocessMaterialID(tmp);
// If Tesseract is checked recognize
if (tmp != null &&
// If bestPoint matches mostCommong point
mostCommon.p.X == bestPoint.p.X && mostCommon.p.Y == bestPoint.p.Y)
{
_ocr.Recognize(tmp);
// Check if (size) result of tesseract is allowed.
if (_ocr.GetCharactors().Length > minimumChars &&
_ocr.GetCharactors().Length < maximumChars)
{
return(_ocr.GetText());
}
}
return(null);
}
public static Tesseract.Charactor[] DoOcr(Image <Gray, byte> img)
{
Tesseract.Charactor[] chars;
string text;
using (var tesseract = new Tesseract("tessdata", "eng", Tesseract.OcrEngineMode.OEM_TESSERACT_CUBE_COMBINED)
)
{
tesseract.Recognize(img);
text = tesseract.GetText();
chars = tesseract.GetCharactors();
}
Console.WriteLine(text);
Console.ReadLine();
return(chars);
}
public static string ReadOCR(Image <Bgr, Byte> srcImage, Rectangle rect)
{
string str_ocr = "";
_ocr = new Tesseract("", "eng", Tesseract.OcrEngineMode.OEM_TESSERACT_ONLY);
// _ocr.SetVariable("tessedit_char_whitelist", "0123456789");
//_ocr.SetVariable("tessedit_char_whitelist", "0123456789ABCDEFGHIJKLMNOPQRSTUVWXYZ");
//_ocr.SetVariable("tessedit_char_whitelist", "abcdefghigklmnopqrstuvwxyz");
_ocr.SetVariable("tessedit_char_whitelist", "0123456789abcdefghigklmnopqrstuvwxyz");
//_ocr.SetVariable("tessedit_char_whitelist", "0123456789abcdefghigklmnopqrstuvwxyz");
try
{
Image <Gray, Byte> Image_Bin = new Image <Gray, Byte>(new Size(rect.Width, rect.Height));
// Image<Gray, Byte> Image_Bin1 = new Image<Gray, Byte>(new Size(rect.Width, rect.Height));
CvInvoke.cvSetImageROI(srcImage, rect);
using (Image <Gray, byte> gray = srcImage.Convert <Gray, Byte>())
{
CvInvoke.cvThreshold(gray, Image_Bin, 0, 255, THRESH.CV_THRESH_OTSU);
// CvInvoke.cvThreshold(Image_Bin, Image_Bin1, 50, 255, THRESH.CV_THRESH_BINARY_INV);
_ocr.Recognize(Image_Bin);
Tesseract.Charactor[] charactors = _ocr.GetCharactors();
ocr_rect = new Rectangle[charactors.Length];
for (int i = 0; i < charactors.Length; i++)
{
ocr_rect[i] = new Rectangle(charactors[i].Region.X + rect.X, charactors[i].Region.Y + rect.Y, charactors[i].Region.Width, charactors[i].Region.Height);
}
str_ocr = _ocr.GetText();
}
Image_Bin.Dispose();
// Image_Bin1.Dispose();
}
catch (Exception exception)
{
//MessageBox.Show(exception.Message);
}
CvInvoke.cvResetImageROI(srcImage);
return(str_ocr);
}
public void TestOCRBgrText()
{
using (Tesseract ocr = GetTesseract())
using (Image <Bgr, Byte> img = new Image <Bgr, byte>(480, 200))
{
ocr.SetVariable("tessedit_char_whitelist", "ABCDEFGHJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz,");
String message = "Hello, World";
CvInvoke.PutText(img, message, new Point(50, 100), CvEnum.FontFace.HersheySimplex, 1.0, new Bgr(Color.Pink).MCvScalar);
//ImageViewer.Show(img);
ocr.Recognize(img);
String messageOcr = ocr.GetText().TrimEnd('\n', '\r'); // remove end of line from ocr-ed text
EmguAssert.AreEqual(message, messageOcr, String.Format("'{0}' is not equal to '{1}'", message, messageOcr));
Tesseract.Charactor[] results = ocr.GetCharactors();
}
}
// read the text from an image
public static String ReadText(Image <Bgr, byte> img)
{
Tesseract _ocr = new Tesseract("", "eng", Tesseract.OcrEngineMode.OEM_TESSERACT_CUBE_COMBINED);
Image <Gray, byte> gray = img.Convert <Gray, Byte>();
{
_ocr.Recognize(gray);
Tesseract.Charactor[] charactors = _ocr.GetCharactors();
/*foreach (Tesseract.Charactor c in charactors)
* {
* img.Draw(c.Region, new Bgr(0,255,0), 1);
* }*/
//imageBox1.Image = image;
//String text = String.Concat( Array.ConvertAll(charactors, delegate(Tesseract.Charactor t) { return t.Text; }) );
//String text = _ocr.GetText();
}
return(_ocr.GetText());
}
private void FindLicensePlate(
Contour <Point> contours, Image <Gray, Byte> gray, Image <Gray, Byte> canny,
List <Image <Gray, Byte> > licensePlateImagesList, List <Image <Gray, Byte> > filteredLicensePlateImagesList, List <MCvBox2D> detectedLicensePlateRegionList,
List <String> licenses)
{
for (; contours != null; contours = contours.HNext)
{
int numberOfChildren = GetNumberOfChildren(contours);
//if it does not contains any children (charactor), it is not a license plate region
if (numberOfChildren == 0)
{
continue;
}
if (contours.Area > 400)
{
if (numberOfChildren < 3)
{
//If the contour has less than 3 children, it is not a license plate (assuming license plate has at least 3 charactor)
//However we should search the children of this contour to see if any of them is a license plate
FindLicensePlate(contours.VNext, gray, canny, licensePlateImagesList, filteredLicensePlateImagesList, detectedLicensePlateRegionList, licenses);
continue;
}
MCvBox2D box = contours.GetMinAreaRect();
if (box.angle < -45.0)
{
float tmp = box.size.Width;
box.size.Width = box.size.Height;
box.size.Height = tmp;
box.angle += 90.0f;
}
else if (box.angle > 45.0)
{
float tmp = box.size.Width;
box.size.Width = box.size.Height;
box.size.Height = tmp;
box.angle -= 90.0f;
}
double whRatio = (double)box.size.Width / box.size.Height;
if (!(3.0 < whRatio && whRatio < 10.0))
//if (!(1.0 < whRatio && whRatio < 2.0))
{ //if the width height ratio is not in the specific range,it is not a license plate
//However we should search the children of this contour to see if any of them is a license plate
Contour <Point> child = contours.VNext;
if (child != null)
{
FindLicensePlate(child, gray, canny, licensePlateImagesList, filteredLicensePlateImagesList, detectedLicensePlateRegionList, licenses);
}
continue;
}
using (Image <Gray, Byte> tmp1 = gray.Copy(box))
//resize the license plate such that the front is ~ 10-12. This size of front results in better accuracy from tesseract
using (Image <Gray, Byte> tmp2 = tmp1.Resize(240, 180, Emgu.CV.CvEnum.INTER.CV_INTER_CUBIC, true))
{
//removes some pixels from the edge
int edgePixelSize = 2;
tmp2.ROI = new Rectangle(new Point(edgePixelSize, edgePixelSize), tmp2.Size - new Size(2 * edgePixelSize, 2 * edgePixelSize));
Image <Gray, Byte> plate = tmp2.Copy();
Image <Gray, Byte> filteredPlate = FilterPlate(plate);
Tesseract.Charactor[] words;
StringBuilder strBuilder = new StringBuilder();
using (Image <Gray, Byte> tmp = filteredPlate.Clone())
{
_ocr.Recognize(tmp);
words = _ocr.GetCharactors();
if (words.Length == 0)
{
continue;
}
for (int i = 0; i < words.Length; i++)
{
strBuilder.Append(words[i].Text);
}
}
licenses.Add(strBuilder.ToString());
licensePlateImagesList.Add(plate);
filteredLicensePlateImagesList.Add(filteredPlate);
detectedLicensePlateRegionList.Add(box);
}
}
}
}
private void ProcessImage(Image <Bgr, byte> image)
{
Stopwatch watch = Stopwatch.StartNew(); // time the detection process
List <Image <Gray, Byte> > licensePlateImagesList = new List <Image <Gray, byte> >();
List <Image <Gray, Byte> > filteredLicensePlateImagesList = new List <Image <Gray, byte> >();
List <MCvBox2D> licenseBoxList = new List <MCvBox2D>();
//List<string> words = _licensePlateDetector.DetectLicensePlate(
// image,
// licensePlateImagesList,
// filteredLicensePlateImagesList,
// licenseBoxList);
var _ocr = new Tesseract(@".\tessdata\", "eng",
Tesseract.OcrEngineMode.OEM_DEFAULT, "1234567890");
List <String> licenses = new List <String>();
using (Image <Gray, byte> gray = image.Convert <Gray, Byte>())
//using (Image<Gray, byte> gray = GetWhitePixelMask(img))
using (Image <Gray, Byte> canny = new Image <Gray, byte>(gray.Size))
using (MemStorage stor = new MemStorage())
{
//CvInvoke.cvCanny(gray, canny, 100, 100, 3);
CvInvoke.cvThreshold(gray, canny, 75, 255, Emgu.CV.CvEnum.THRESH.CV_THRESH_BINARY);
var bibs = new List <Image <Gray, Byte> >();
var bibsreg = new List <Rectangle>();
//imageBox1.Image = gray;
var faces = DF(image);
Point startPoint = new Point(10, 10);
this.Text = "";
foreach (Rectangle face in faces)
{
canny.Draw(face, new Gray(1), 2);
var r = new Rectangle(face.X - face.Width / 2, face.Y + face.Height * 2,
face.Width * 2, face.Height * 3);
bibsreg.Add(r);
CvInvoke.cvSetImageROI(canny, r);
var ni = canny.Copy();
_ocr.Recognize(ni);
var words = _ocr.GetCharactors();
CvInvoke.cvResetImageROI(canny);
this.Text += " bib:" + _ocr.GetText().Replace(" ", string.Empty);
//AddLabelAndImage(ref startPoint, _ocr.GetText(), ni);
}
foreach (Rectangle r in bibsreg)
{
canny.Draw(r, new Gray(1), 10);
}
imageBox1.Image = canny;
//_ocr.Recognize(canny);
//var words = _ocr.GetCharactors();
//this.Text = _ocr.GetText();
//Contour<Point> contours = canny.FindContours(
// Emgu.CV.CvEnum.CHAIN_APPROX_METHOD.CV_CHAIN_APPROX_SIMPLE,
// Emgu.CV.CvEnum.RETR_TYPE.CV_RETR_TREE,
// stor);
//FindLicensePlate(contours, gray, canny, licensePlateImagesList, filteredLicensePlateImagesList, detectedLicensePlateRegionList, licenses);
}
watch.Stop(); //stop the timer
processTimeLabel.Text = String.Format("License Plate Recognition time: {0} milli-seconds", watch.Elapsed.TotalMilliseconds);
//panel1.Controls.Clear();
//Point startPoint = new Point(10, 10);
//for (int i = 0; i < words.Count; i++)
//{
// AddLabelAndImage(
// ref startPoint,
// String.Format("License: {0}", words[i]),
// licensePlateImagesList[i].ConcateVertical(filteredLicensePlateImagesList[i]));
// image.Draw(licenseBoxList[i], new Bgr(Color.Red), 2);
//}
//imageBox1.Image = image;
}
/// <summary>
/// Processed Image is the one with some rectangles
/// </summary>
/// <param name="inputImage"></param>
/// <param name="ProcessedImage"></param>
/// <returns></returns>
public static string OCRDetect(Image <Bgr, byte> inputImage, out Image <Bgr, byte> ProcessedImage)
{
ProcessedImage = inputImage;
Bgr drawColor = new Bgr(Color.Blue);
Image <Bgr, byte> localImage = inputImage.Copy();
try
{
using (Image <Gray, byte> gray = localImage.Convert <Gray, Byte>())
{
_ocr.Recognize(gray);
Tesseract.Charactor[] charactors = _ocr.GetCharactors();
/*
* // This will filter the charactors that are too small. But noise is going to affect the accuracy already
* List<Tesseract.Charactor> charactorsList = new List<Tesseract.Charactor>();
* for (int i = 0; i < charactors.Length; i++)
* {
* if (charactors[i].Region.Width * charactors[i].Region.Height < 100) //Too Small
* {
* localImage.Draw(charactors[i].Region, new Bgr(Color.Red), 1);
* }
* else
* {
* localImage.Draw(charactors[i].Region, drawColor, 1);
* charactorsList.Add(charactors[i]);
* }
* }
*
* Tesseract.Charactor[] newCharactors = new Tesseract.Charactor[charactorsList.Count];
* for (int i = 0; i < newCharactors.Length; i++)
* {
* newCharactors[i] = charactorsList[i];
* }
*
*/
foreach (Tesseract.Charactor c in charactors)
{
if (c.Region.Width * c.Region.Height < 200)
{
localImage.Draw(c.Region, new Bgr(Color.Red), 1);
//ShapeNDraw.drawString(c.Cost.ToString(), cleanedInputImage, new Point(c.Region.X, c.Region.Y), 0.3, Color.Red);
}
else
{
localImage.Draw(c.Region, drawColor, 1);
//ShapeNDraw.drawString(c.Cost.ToString(), cleanedInputImage, new Point(c.Region.X, c.Region.Y), 0.3, Color.Red);
}
}
ProcessedImage = localImage;
//string text = String.Concat( Array.ConvertAll(newCharactors, delegate(Tesseract.Charactor t) { return t.Text; }) );
string text = _ocr.GetText();
return(text);
}
}
catch (Exception exception)
{
return("NULL");
}
}
