///''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''
public static void preprocess(Mat imgOriginal, ref Mat imgGrayscale, ref Mat imgThresh)
{
imgGrayscale = extractValue(imgOriginal);
//extract value channel only from original image to get imgGrayscale
Mat imgMaxContrastGrayscale = imgGrayscale;
//Mat imgMaxContrastGrayscale = maximizeContrast(imgGrayscale);
//maximize contrast with top hat and black hat
Mat imgBlurred = new Mat();
CvInvoke.GaussianBlur(imgMaxContrastGrayscale, imgBlurred, new Size(GAUSSIAN_BLUR_FILTER_SIZE, GAUSSIAN_BLUR_FILTER_SIZE), 0.6);
//gaussian blur
//adaptive threshold to get imgThresh
CvInvoke.AdaptiveThreshold(imgBlurred, imgThresh, 255.0, AdaptiveThresholdType.GaussianC, ThresholdType.BinaryInv, ADAPTIVE_THRESH_BLOCK_SIZE, ADAPTIVE_THRESH_WEIGHT);
MCvScalar tempVal = CvInvoke.Mean(imgBlurred);
double average = tempVal.V0;
CvInvoke.Threshold(imgBlurred, imgThresh, 0, 255.0, Emgu.CV.CvEnum.ThresholdType.Otsu);
CvInvoke.Erode(imgThresh, imgThresh, null, Point.Empty, 1, BorderType.Default, new MCvScalar(0));
CvInvoke.Dilate(imgThresh, imgThresh, null, Point.Empty, 1, BorderType.Default, new MCvScalar(0));
CvInvoke.BitwiseNot(imgThresh, imgThresh);
}
private void BubbleDetectBtn_Click(object sender, EventArgs e)
{
//Applying Operations on transformed Image
transformedImage = transformedImage.Resize(400, 400, Emgu.CV.CvEnum.Inter.Linear);
Image <Bgr, byte> transCopy = transformedImage.Copy();
Emgu.CV.Util.VectorOfVectorOfPoint qtnVect = new Emgu.CV.Util.VectorOfVectorOfPoint();
Image <Gray, byte> qtnGray = transCopy.Convert <Gray, byte>();
Image <Gray, byte> copyG = qtnGray.Copy();
CvInvoke.GaussianBlur(qtnGray, qtnGray, new Size(5, 5), 0);
CvInvoke.AdaptiveThreshold(qtnGray, qtnGray, 255, Emgu.CV.CvEnum.AdaptiveThresholdType.GaussianC, Emgu.CV.CvEnum.ThresholdType.Binary, 55, 9);
CvInvoke.BitwiseNot(qtnGray, qtnGray);
CvInvoke.FindContours(qtnGray, qtnVect, null, Emgu.CV.CvEnum.RetrType.External, Emgu.CV.CvEnum.ChainApproxMethod.ChainApproxSimple, default);
//CIRCLE METHOD
List <CircleF> circList = new List <CircleF>();
Emgu.CV.Util.VectorOfVectorOfPoint test = new Emgu.CV.Util.VectorOfVectorOfPoint();
Emgu.CV.Util.VectorOfPoint qtnApprox = new Emgu.CV.Util.VectorOfPoint();
Dictionary <int, double> qtnDict = new Dictionary <int, double>();
if (qtnVect.Size > 0)
{
for (int i = 0; i < qtnVect.Size; i++)
{
double area = CvInvoke.ContourArea(qtnVect[i]);
if (area > 70)
{
qtnDict.Add(i, area);
}
}
var item = qtnDict.OrderByDescending(v => v.Value); //.Take(1);
Emgu.CV.Util.VectorOfPoint approxList = new Emgu.CV.Util.VectorOfPoint();
foreach (var it in item)
{
int key = Convert.ToInt32(it.Key.ToString());
double peri = CvInvoke.ArcLength(qtnVect[key], true);
CvInvoke.ApproxPolyDP(qtnVect[key], qtnApprox, 0.02 * peri, true);
if (qtnApprox.Size == 0)
{
}
else if (qtnApprox.Size > 6)
{
CircleF circle = CvInvoke.MinEnclosingCircle(qtnVect[key]);
Point centre = new Point();
centre.X = (int)circle.Center.X;
centre.Y = (int)circle.Center.Y;
CvInvoke.Circle(transformedImage, centre, (int)circle.Radius, new MCvScalar(0, 255, 0), 2, Emgu.CV.CvEnum.LineType.Filled, 0);
//break;
}
}
MessageBox.Show("Bubbles Detected");
bubbleImage.Image = transformedImage;
}
}
/// <summary>
/// Pass the image through multiple filters and sort contours
/// </summary>
/// <param name="img">The image that will be proccessed</param>
/// <returns>A list of Mat ROIs</returns>
private static List <Mat> ImageProccessing(Mat img)
{
//Resize the image for better uniformitty throughout the code
CvInvoke.Resize(img, img, new Size(700, 500));
Mat imgClone = img.Clone();
//Convert the image to grayscale
CvInvoke.CvtColor(img, img, ColorConversion.Bgr2Gray);
//Blur the image
CvInvoke.GaussianBlur(img, img, new Size(5, 5), 8, 8);
//Threshold the image
CvInvoke.AdaptiveThreshold(img, img, 30, AdaptiveThresholdType.GaussianC, ThresholdType.Binary, 5, 6);
//Canny the image
CvInvoke.Canny(img, img, 8, 8);
//Dilate the canny image
CvInvoke.Dilate(img, img, null, new Point(-1, -1), 8, BorderType.Constant, new MCvScalar(0, 255, 255));
//Filter the contours to only find relevent ones
List <Mat> foundOutput = FindandFilterContours(imgClone, img);
return(foundOutput);
}
public void GridDetection()
{
// convert to gray-scaler image
Mat image = originalImage.Mat.Clone();
// blur the image
CvInvoke.GaussianBlur(image, image, new Size(11, 11), 0);
// threshold the image
CvInvoke.AdaptiveThreshold(image, image, 255, AdaptiveThresholdType.MeanC, ThresholdType.Binary, 5, 2);
CvInvoke.BitwiseNot(image, image);
Mat kernel = new Mat(new Size(3, 3), DepthType.Cv8U, 1);
Marshal.Copy(new byte[] { 0, 1, 0, 1, 1, 1, 0, 1, 0 }, 0, kernel.DataPointer, 9);
CvInvoke.Dilate(image, image, kernel, new Point(-1, -1), 1, BorderType.Default, new MCvScalar(255));
FindOuterGridByFloorFill(image);
CvInvoke.Erode(image, image, kernel, new Point(-1, -1), 1, BorderType.Default, new MCvScalar(255));
ImageShowCase.ShowImage(image, "biggest blob");
VectorOfPointF lines = new VectorOfPointF();
CvInvoke.HoughLines(image, lines, 1, Math.PI / 180, 200);
// merging lines
PointF[] linesArray = lines.ToArray();
//MergeLines(linesArray, image);
lines = RemoveUnusedLine(linesArray);
Mat harrisResponse = new Mat(image.Size, DepthType.Cv8U, 1);
CvInvoke.CornerHarris(image, harrisResponse, 5);
DrawLines(lines.ToArray(), image);
ImageShowCase.ShowImage(image, "corners");
}
private void PreProcessImage()
{
if (ImagesPaths.Count < 1)
{
MessageBox.Show("No Image To Process!");
return;
}
for (int index = 0; index <= nImages - 1; index++)
{
Mat GrayImage = ImagesReadCV[index].Clone();
Mat BlurredImage = GrayImage.Clone();
Mat AdaptThresh = BlurredImage.Clone();
if (GrayImage.NumberOfChannels != 3)
{
MessageBox.Show(@"Invalid Number of Channels within the Image!\nExpect 3 but got" + GrayImage.NumberOfChannels);
return;
}
CvInvoke.CvtColor(ImagesReadCV[index], GrayImage, ColorConversion.Bgr2Gray);
CvInvoke.GaussianBlur(GrayImage, BlurredImage, new Size(3, 3), 1);
CvInvoke.AdaptiveThreshold(BlurredImage, AdaptThresh, 255, AdaptiveThresholdType.GaussianC, ThresholdType.BinaryInv, 11, 2);
ImagesReadCV[index] = AdaptThresh;
}
MessageBox.Show("Done!");
ReloadImageBox();
}
private void HandleGrab(object sender, EventArgs e)
{
Mat image = new Mat();
if (capture.IsOpened)
{
capture.Retrieve(image);
}
if (image.IsEmpty)
{
return;
}
Mat grayImg = image.Clone();
CvInvoke.CvtColor(image, grayImg, ColorConversion.Bgr2Gray);
CvInvoke.AdaptiveThreshold(grayImg, grayImg, 255, AdaptiveThresholdType.MeanC, ThresholdType.BinaryInv, 21, 11);
VectorOfInt ids = new VectorOfInt();
VectorOfVectorOfPointF corners = new VectorOfVectorOfPointF();
VectorOfVectorOfPointF rejected = new VectorOfVectorOfPointF();
ArucoInvoke.DetectMarkers(image, dico, corners, ids, arucoParam, rejected);
if (ids.Size > 0)
{
ArucoInvoke.DrawDetectedMarkers(image, corners, ids, new MCvScalar(0, 0, 255));
}
CvInvoke.Imshow("Original", image);
CvInvoke.Imshow("Gray", grayImg);
}
public static Mat PreprocessImageForTesseract(Mat img)
{
int scalePercent = 18;
int newWidth = (int)img.Width * scalePercent / 100;
int newHeight = (int)img.Height * scalePercent / 100;
CvInvoke.Resize(img, img, new System.Drawing.Size(newWidth, newHeight), interpolation: Inter.Area);
img = ImageProcessor.ApplyBlur(img, 0, 3);
Mat output = new Mat(img.Size, DepthType.Cv8U, 3);
CvInvoke.CvtColor(img, img, ColorConversion.Bgr2Gray);
//CvInvoke.EqualizeHist(img, img);
CvInvoke.BitwiseNot(img, img);
//img = ImageProcessor.CannyEdgeDetection(img, 20, 20);
//img = ImageProcessor.ApplyErosion(img, 3);
//CvInvoke.GaussianBlur(img, img, new System.Drawing.Size(3, 3), 0);
CvInvoke.AdaptiveThreshold(img, img, 255, AdaptiveThresholdType.GaussianC, ThresholdType.Binary, 11, 2);
CvInvoke.Threshold(img, output, 0, 255, ThresholdType.Otsu);//double ret =
//output = ImageProcessor.ApplyErosion(output, 3);
//CvInvoke.Threshold(output, output, ret, 255, ThresholdType.Binary);
var kernel = CvInvoke.GetStructuringElement(Emgu.CV.CvEnum.ElementShape.Rectangle, new System.Drawing.Size(2, 2), new System.Drawing.Point(-1, -1));
CvInvoke.Dilate(output, output, kernel, new System.Drawing.Point(-1, -1), 2, Emgu.CV.CvEnum.BorderType.Constant, default(MCvScalar));
//output = ImageProcessor.ApplyDilation(output, 7);
//CvInvoke.Invert()
return(output);
}
private void ImageThreshold()
{
if (grayImg != null)
{
try
{
if (cboAdapEn.Checked)
{
CvInvoke.AdaptiveThreshold(grayImg, binImg, 255, (AdaptiveThresholdType)cboAdapType.SelectedItem,
(ThresholdType)comboBox1.SelectedItem,
Convert.ToInt32((2 * tbrMSize.Value) + 1), //require only Odd number and > 0
tbrMValue.Value);
}
else
{
CvInvoke.Threshold(grayImg, binImg, th, 255, (Emgu.CV.CvEnum.ThresholdType)comboBox1.SelectedItem);
}
}
catch (Exception ex)
{
MessageBox.Show(ex.Message);
return;
}
pboGray.Image = binImg.ToBitmap(pboGray.Width, pboGray.Height);
}
}
public static Image <Gray, byte> TresholdingAdative(Image <Gray, byte> image, int blockSize = 11)
{
var tres = image.CopyBlank();
CvInvoke.AdaptiveThreshold(image, tres, 255, AdaptiveThresholdType.GaussianC, ThresholdType.Binary, blockSize, -1);
return(tres);
}
public void HandleWebcamQueryFrame(object sender, EventArgs e)
{
if (webcam.IsOpened)
{
webcam.Retrieve(image);
}
if (image.IsEmpty)
{
return;
}
imageGray = image.Clone();
CvInvoke.CvtColor(image, imageGray, ColorConversion.Bgr2Gray);
if (imageGray.IsEmpty)
{
return;
}
frontFaces = frontFaceCascadeClassifier.DetectMultiScale(image: imageGray, scaleFactor: 1.1, minNeighbors: 5, minSize: new Size(MIN_FACE_SIZE, MIN_FACE_SIZE), maxSize: new Size(MAX_FACE_SIZE, MAX_FACE_SIZE));
Debug.Log(frontFaces.Length.ToString());
for (int i = 0; i < frontFaces.Length; i++)
{
CvInvoke.Rectangle(image, frontFaces[i], new MCvScalar(0, 180, 0), 0);
Debug.Log("i: " + i.ToString());
}
//Nouvelle matrice qui focus sur le premier visage
if (frontFaces.Length > 0)
{
image = new Mat(image, frontFaces[0]);
}
DisplayFrame(image);
//Seuillage adaptatif
Mat hierarchy = new Mat();
CvInvoke.AdaptiveThreshold(imageGray, imageGray, maxValue, AdaptiveThresholdType.MeanC, ThresholdType.Binary, blockSize, diviser);
CvInvoke.FindContours(imageGray, allContours, hierarchy, RetrType.List, ChainApproxMethod.ChainApproxNone);
desiredContours.Clear();
for (int i = 0; i < allContours.Size; i++)
{
if (CvInvoke.ContourArea(allContours[i]) > contourSizeMin && CvInvoke.ContourArea(allContours[i]) < contourSizeMax)
{
desiredContours.Push(allContours[i]);
}
}
CvInvoke.DrawContours(image, desiredContours, -1, new MCvScalar(200, 100, 200), 2);
//RotatedRect rotatedRect;
//rotatedRect = CvInvoke.MinAreaRect(biggestContour);
//rotatedRect.GetVertices();
CvInvoke.Imshow("Webcam view Normal", image);
CvInvoke.Imshow("Webcam view Gray", imageGray);
}
private Image <Gray, byte> preprocess(Image <Gray, byte> input)
{
Image <Gray, byte> copy = input.Copy();
CvInvoke.Blur(copy, copy, new Size(5, 5), new Point(-1, -1));
CvInvoke.AdaptiveThreshold(copy, copy, 255, Emgu.CV.CvEnum.AdaptiveThresholdType.GaussianC, Emgu.CV.CvEnum.ThresholdType.BinaryInv, 11, 2);
return(copy);
}
public static Image <Gray, Byte> GetAdaptiveThresholdedFrame(Image <Bgr, Byte> frame)
{
var result = new Mat();
CvInvoke.CvtColor(frame, result, ColorConversion.Bgr2Gray);
CvInvoke.AdaptiveThreshold(result, result, 255, AdaptiveThresholdType.GaussianC, ThresholdType.Binary, 15, -10);
return(result.ToImage <Gray, Byte>());
}
private void preencherImagemBinariaSemPlanoDeFundo()
{
mCopiaImagemPlanoDeFundo = mImagemDoPlanoDeFundo.Clone();
CvInvoke.AbsDiff(mImagemColorida, mCopiaImagemPlanoDeFundo, mImagemSemPlanoDeFundo);
CvInvoke.CvtColor(mImagemSemPlanoDeFundo, mImagemCinzaSemPlanoDeFundo, ColorConversion.Rgb2Gray);
// CvInvoke.Threshold(mImagemCinzaSemPlanoDeFundo, mImagemBinariaSemPlanoDeFundo, ParametrosConstantes.LimiarTransformacaoParaCinza,
//ParametrosConstantes.MaximoLimiarTransformacaoParaCinza, ThresholdType.Binary);
CvInvoke.AdaptiveThreshold(mImagemCinzaSemPlanoDeFundo, mImagemBinariaSemPlanoDeFundo, ParametrosConstantes.MaximoLimiarTransformacaoParaCinza,
AdaptiveThresholdType.GaussianC, ThresholdType.Binary, 11, 3);
}
private void button1_Click(object sender, EventArgs e)
{
Image <Gray, byte> img = new Bitmap(PbSegmentation.Image).ToImage <Gray, byte>();
CvInvoke.AdaptiveThreshold(img, img, 255,
Emgu.CV.CvEnum.AdaptiveThresholdType.GaussianC,
Emgu.CV.CvEnum.ThresholdType.Binary, 11, 5);
PbSegmentation.Image = img.ToBitmap();
Tools.Histogram(ChartSegmentation, (Bitmap)PbSegmentation.Image);
}
private void button3_Click(object sender, EventArgs e)
{
Image <Gray, Byte> imgbuff = new Image <Gray, byte>(imagetest.Width, imagetest.Height);
CvInvoke.AdaptiveThreshold(imagetest, imgbuff, 255, AdaptiveThresholdType.MeanC, ThresholdType.BinaryInv, 3, 0);
imagetest = imgbuff;
imageBox2.Image = imagetest;
}
private string[] ProcessSudokuImg(string fileName) // --> Wydzielic do innej klasy!
{
int height = 450;
int width = 450;
string[] digits = new string[81];
Image <Bgr, byte> image = new Image <Bgr, byte>(fileName);
image = image.Resize(width, height, Emgu.CV.CvEnum.Inter.Linear);
Image <Gray, byte> grayImage = image.Convert <Gray, byte>();
Image <Gray, byte> buffer = grayImage.Copy();
CvInvoke.GaussianBlur(grayImage, buffer, new System.Drawing.Size(5, 5), 1);
grayImage = buffer;
CvInvoke.AdaptiveThreshold(grayImage, buffer, 255, Emgu.CV.CvEnum.AdaptiveThresholdType.GaussianC, Emgu.CV.CvEnum.ThresholdType.Binary, 5, 2);
grayImage = buffer;
// Split image into 81 parts
Image <Gray, byte>[] fields = new Image <Gray, byte> [81];
for (int i = 0; i < 9; i++)
{
for (int j = 0; j < 9; j++)
{
int border = 5;
System.Drawing.Rectangle rect = new System.Drawing.Rectangle(i * (width / 9) + border, j * (height / 9) + border, (width / 9) - 2 * border, (height / 9) - 2 * border);
grayImage.ROI = rect;
var index = i * 9 + j;
fields[index] = grayImage.CopyBlank();
grayImage.CopyTo(fields[index]);
grayImage.ROI = System.Drawing.Rectangle.Empty;
}
}
// Recognize digits
using (TesseractEngine engine = new TesseractEngine(@"./tessdata", "eng", EngineMode.Default))
{
engine.SetVariable("tessedit_char_whitelist", "0123456789");
int i = 0; //iterator
foreach (var field in fields)
{
Page page = engine.Process(field.ToBitmap(), PageSegMode.SingleChar);
string result = page.GetText();
page.Dispose();
digits[i++] = result.Trim();
field.Dispose();
}
}
image.Dispose(); grayImage.Dispose(); buffer.Dispose();
return(digits);
}
public static void preprocess(Mat imgOriginal, Mat imgGrayscale, Mat imgThresh)//przerobic na kontruktor??????
{
imgGrayscale = extractValue(imgOriginal);
Mat imgMaxContrastGrayscale = maximizeContrast(imgGrayscale);
Mat imgBlurred = new Mat();
CvInvoke.GaussianBlur(imgMaxContrastGrayscale, imgBlurred, new Size(GAUSSIAN_BLUR_FILTER_SIZE, GAUSSIAN_BLUR_FILTER_SIZE), 0);
CvInvoke.AdaptiveThreshold(imgBlurred, imgThresh, 255.0, AdaptiveThresholdType.GaussianC, ThresholdType.BinaryInv, ADAPTIVE_THRESH_BLOCK_SIZE, ADAPTIVE_THRESH_WEIGHT);
}
/// <summary>
/// Processes the infrared frame:
/// 1. Thesholds the infrared image
/// 2. Opens the thesholded image
/// 3. Tracks refletive markers in the thresholded image.
/// 4. Show infrared/thresholded image if mainwindow is present
/// </summary>
/// <param name="infraredFrame"> the InfraredFrame image </param>
/// <param name="infraredFrameDataSize">Size of the InfraredFrame image data</param>
private void ProcessIRFrame(Mat infraredFrameOrg, FrameDimension infraredFrameDimension, Mat depthFrame)
{
// init threshold image variable
// Image<Gray, Byte> thresholdImg = new Image<Gray, Byte>(infraredFrameDimension.Width, infraredFrameDimension.Height);
using (Mat thresholdImg = new Mat(),
infraredFrameROI = new Mat(infraredFrameOrg, mask))
{
CvInvoke.Normalize(infraredFrameROI, thresholdImg, 0, 255, NormType.MinMax, DepthType.Cv8U);
CvInvoke.AdaptiveThreshold(thresholdImg, thresholdImg, 255, AdaptiveThresholdType.GaussianC, ThresholdType.Binary, 13, -20);
// perform opening
CvInvoke.MorphologyEx(thresholdImg, thresholdImg, MorphOp.Dilate, kernel, new System.Drawing.Point(-1, -1), 2, BorderType.Constant, new MCvScalar(1.0));
CvInvoke.MorphologyEx(thresholdImg, thresholdImg, MorphOp.Erode, kernel, new System.Drawing.Point(-1, -1), 1, BorderType.Constant, new MCvScalar(1.0));
// find controids of reflective surfaces and mark them on the image
double[][] centroidPoints = GetvisibleData(thresholdImg, depthFrame);
TrackedData(centroidPoints);
// only generate writeable bitmap if the mainwindow is shown
if (this.showWindow)
{
// copy the processed image back into a writeable bitmap and dispose the EMGU image
if (thresholdedClicked)
{
using (DrawTrackedData(thresholdImg))
{
SetThresholdedInfraredImage(thresholdImg, infraredFrameDimension);
}
}
else
{
using (Mat colImg = DrawTrackedData(infraredFrameOrg))
{
// CvInvoke.Normalize(colImg, colImg, 0, 255, NormType.MinMax, DepthType.Cv8U);
SetInfraredImage(colImg, infraredFrameDimension);
}
}
}
// cleanup
}
}
public static Bitmap getBinarizedBitmap(Bitmap bitmap)
{
var image = new Image <Bgr, byte>(bitmap);
var uimage = new UMat();
var pyrDown = new UMat();
var imageBynarize = image.Convert <Gray, Byte>();
CvInvoke.CvtColor(image, uimage, ColorConversion.Bgr2Gray);
CvInvoke.PyrDown(uimage, pyrDown);
CvInvoke.PyrUp(pyrDown, uimage);
CvInvoke.AdaptiveThreshold(imageBynarize, imageBynarize, 255, AdaptiveThresholdType.GaussianC, ThresholdType.Binary, 255, 16);
return(imageBynarize.ToBitmap(bitmap.Width, bitmap.Height));
}
private void GetImageEdges()
{
CvInvoke.MedianBlur(workImage, workImage, 3);
// ThresholdType.Binary, 11, 2
CvInvoke.AdaptiveThreshold(workImage, workImage, 255, AdaptiveThresholdType.GaussianC, ThresholdType.BinaryInv, 101, 3);
CvInvoke.MedianBlur(workImage, workImage, 3);
workImage = workImage.MorphologyEx(MorphOp.Dilate, kernel, new Point(-1, -1), 1, BorderType.Default, new MCvScalar(1.0));
CvInvoke.Canny(workImage, workImage, 80.0, 120.0);
}
// Update is called once per frame
void Update()
{
fluxVideo.Grab();
Mat grey = new Mat();
CvInvoke.CvtColor(image, grey, ColorConversion.Bgr2Gray);
CvInvoke.AdaptiveThreshold(grey, grey, 255, AdaptiveThresholdType.GaussianC, ThresholdType.BinaryInv, 21, 11);
CvInvoke.FindContours(grey, contours, m, Emgu.CV.CvEnum.RetrType.External, Emgu.CV.CvEnum.ChainApproxMethod.ChainApproxSimple);
for (int i = 0; i < contours.Size; i++)
{
double perimeter = CvInvoke.ArcLength(contours[i], true);
CvInvoke.ApproxPolyDP(contours[i], approx, 0.04 * perimeter, true);
if (approx.Size == 4)
{
if (CvInvoke.ContourArea(contours[i]) > 300)
{
var rect = CvInvoke.BoundingRectangle(approx);
if (rect.Height > 0.95 * rect.Width || rect.Height < 0.95 * rect.Width)
{
candidates.Push(approx);
CvInvoke.DrawContours(image, contours, i, new MCvScalar(0, 255, 0), 4);
CvInvoke.Rectangle(image, rect, new MCvScalar(255, 0, 0));
}
}
}
}
for (int i = 0; i < candidates.Size; i++)
{
System.Drawing.PointF[] pts = new System.Drawing.PointF[4];
pts[0] = new System.Drawing.PointF(0, 0);
pts[1] = new System.Drawing.PointF(64 - 1, 0);
pts[2] = new System.Drawing.PointF(64 - 1, 64 - 1);
pts[3] = new System.Drawing.PointF(0, 64 - 1);
VectorOfPointF perfect = new VectorOfPointF(pts);
System.Drawing.PointF[] sample_pts = new System.Drawing.PointF[4];
for (int ii = 0; ii < 4; ii++)
{
sample_pts[ii] = new System.Drawing.PointF(candidates[i][ii].X, candidates[i][ii].Y);
}
VectorOfPointF sample = new VectorOfPointF(sample_pts);
var tf = CvInvoke.GetPerspectiveTransform(sample, perfect);
Mat warped = new Mat();
CvInvoke.WarpPerspective(image, warped, tf, new System.Drawing.Size(64, 64));
CvInvoke.Imshow("yo", warped);
}
CvInvoke.WaitKey(24);
}
public static Image <Gray, byte> AdaptiveThreshold
(
this Image <Gray, byte> inImage
, double maxVal = 250
, AdaptiveThresholdType adaptiveThresholdType = AdaptiveThresholdType.MeanC
, ThresholdType thresholdType = ThresholdType.BinaryInv
, int blockSize = 39
, double param1 = 4)
{
var outImage = inImage.Copy();
CvInvoke.AdaptiveThreshold
(GaussBlur(inImage), outImage, maxVal, adaptiveThresholdType, thresholdType, blockSize, param1);
return(outImage);
}
private void Button3_Click(object sender, EventArgs e)
{
imgGray = img.Convert <Gray, Byte>();
//Converted to blurred
CvInvoke.GaussianBlur(imgGray, imgGray, new Size(5, 5), 0);
// using adaptive threshhold
CvInvoke.AdaptiveThreshold(imgGray, imgGray, 255, Emgu.CV.CvEnum.AdaptiveThresholdType.GaussianC, Emgu.CV.CvEnum.ThresholdType.BinaryInv, 75, 10);
CvInvoke.Canny(imgGray, cannyImage, 75, 200);
cannyImage.ConvertTo(imgGray, Emgu.CV.CvEnum.DepthType.Default, -1, 0);
Emgu.CV.Util.VectorOfVectorOfPoint vector = new Emgu.CV.Util.VectorOfVectorOfPoint();
CvInvoke.FindContours(cannyImage, vector, null, Emgu.CV.CvEnum.RetrType.External, Emgu.CV.CvEnum.ChainApproxMethod.ChainApproxSimple);
CvInvoke.DrawContours(img, vector, -1, new MCvScalar(240, 0, 159), 3);
MessageBox.Show("Question Part Detected");
sheetDetectImage.Image = img;
}
private static Image <Gray, byte> getBinarizedImage(Bitmap bitmap)
{
var image = new Image <Bgr, byte>(bitmap);
var uimage = new UMat();
var pyrDown = new UMat();
CvInvoke.CvtColor(image, uimage, ColorConversion.Bgr2Gray);
CvInvoke.PyrDown(uimage, pyrDown);
CvInvoke.PyrUp(pyrDown, uimage);
var imageBynarize = image.Convert <Gray, Byte>().PyrUp().PyrDown();
CvInvoke.AdaptiveThreshold(imageBynarize, imageBynarize, 255, AdaptiveThresholdType.MeanC, ThresholdType.Binary, 115, 4);
CvInvoke.Threshold(image, image, 200, 255, ThresholdType.Otsu);
return(imageBynarize);
}
private void binarizeToolStripMenuItem_Click(object sender, EventArgs e)
{
imgGray = imgInput.Convert <Gray, byte>();
pictureBox1.Image = imgGray.ToBitmap();
// Binarization
imgBinarized = new Image <Gray, byte>(imgGray.Width, imgGray.Height, new Gray(0));
//CvInvoke.Threshold(imgGray, imgBinarized, 50, 255, Emgu.CV.CvEnum.ThresholdType.Binary);
//pictureBox2.Image = imgBinarized.ToBitmap();
// adaptive threshold
CvInvoke.AdaptiveThreshold(imgGray, imgBinarized, 255, Emgu.CV.CvEnum.AdaptiveThresholdType.GaussianC, Emgu.CV.CvEnum.ThresholdType.Binary, 5, 0.0);
pictureBox2.Image = imgBinarized.ToBitmap();
}
private static Mat Preprocess(Mat image, bool isPerfectShape)
{
Mat outerBox = new Mat(image.Size, DepthType.Cv8U, 3);
CvInvoke.GaussianBlur(image, image, new Size(7, 7), 0);
CvInvoke.AdaptiveThreshold(image, outerBox, 255, AdaptiveThresholdType.MeanC, ThresholdType.Binary, 5, 2);
CvInvoke.BitwiseNot(outerBox, outerBox);
if (!isPerfectShape || true)
{
var element = CvInvoke.GetStructuringElement(ElementShape.Cross, new Size(3, 3), new Point(-1, -1));
CvInvoke.Dilate(outerBox, outerBox, element, new Point(-1, -1), 1, BorderType.Default, default(MCvScalar));
CvInvoke.Erode(outerBox, outerBox, element, new Point(-1, -1), 1, BorderType.Default, default(MCvScalar));
}
return(outerBox);
}
///''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''
public static void preprocess(Mat imgOriginal, ref Mat imgGrayscale, ref Mat imgThresh)
{
imgGrayscale = extractValue(imgOriginal);
//extract value channel only from original image to get imgGrayscale
Mat imgMaxContrastGrayscale = maximizeContrast(imgGrayscale);
//maximize contrast with top hat and black hat
Mat imgBlurred = new Mat();
CvInvoke.GaussianBlur(imgMaxContrastGrayscale, imgBlurred, new Size(GAUSSIAN_BLUR_FILTER_SIZE, GAUSSIAN_BLUR_FILTER_SIZE), 0);
//gaussian blur
//adaptive threshold to get imgThresh
CvInvoke.AdaptiveThreshold(imgBlurred, imgThresh, 255.0, AdaptiveThresholdType.GaussianC, ThresholdType.BinaryInv, ADAPTIVE_THRESH_BLOCK_SIZE, ADAPTIVE_THRESH_WEIGHT);
}
private void AdaptiveThresholdingBarButtonItem_ItemClick(object sender, DevExpress.XtraBars.ItemClickEventArgs e)
{
Directory.CreateDirectory(@"D:\Eighth Semester\HandVeinPattern\\RuntimeDirectory");
Details.adaptivethreshold = new Mat();
CvInvoke.AdaptiveThreshold(Details.whitebalance, Details.adaptivethreshold, 255, AdaptiveThresholdType.GaussianC, ThresholdType.Binary, 11, -3);
Step3PictureEdit.Image = Details.adaptivethreshold.Bitmap;
CvInvoke.Imwrite(@"D:\Eighth Semester\HandVeinPattern\RuntimeDirectory\AdaptiveThreshold.jpg", Details.adaptivethreshold);
ImageProcessingProgressBarControl.PerformStep();
ImageProcessingProgressBarControl.Update();
}
public Bitmap AdaptiveBinarization(string imageSavePath, int sizeOfNbPixels, int subFromMean)
{
Mat img = CvInvoke.Imread(imageSavePath, 0);
Mat output = new Mat();
CvInvoke.AdaptiveThreshold(img, output, 255,
AdaptiveThresholdType.GaussianC, ThresholdType.Binary, sizeOfNbPixels, subFromMean);
CvInvoke.Imshow("src", img);
CvInvoke.Imshow("Gaussian", output);
Bitmap newMap = output.ToBitmap();
newMap.Save("./Tesseract Ocr/testImage111.png");
return(newMap);
}
/// <summary>
/// Pass the image through multiple filters and sort contours
/// </summary>
/// <param name="img">The image that will be proccessed</param>
/// <returns>A list of Mat ROIs</returns>
private static Mat ImageProccessing(Mat img)
{
//Resize the image for better uniformitty throughout the code
CvInvoke.Resize(img, img, new System.Drawing.Size(600, 800));
Mat imgClone = img.Clone();
//Convert the image to grayscale
CvInvoke.CvtColor(img, img, ColorConversion.Bgr2Gray);
//Blur the image
CvInvoke.GaussianBlur(img, img, new System.Drawing.Size(3, 3), 4, 4);
CvInvoke.Imshow("GaussianBlur", img);
CvInvoke.WaitKey(0);
//Threshold the image
CvInvoke.AdaptiveThreshold(img, img, 100, AdaptiveThresholdType.GaussianC, ThresholdType.BinaryInv, 5, 6);
CvInvoke.Imshow("Thereshold", img);
CvInvoke.WaitKey(0);
/* //Canny the image
* CvInvoke.Canny(img, img, 75, 100);
*
* CvInvoke.Imshow("Canny", img);
* CvInvoke.WaitKey(0);*/
/* //Dilate the canny image
* CvInvoke.Dilate(img, img, null, new System.Drawing.Point(-1, -1), 8, BorderType.Constant, new MCvScalar(0, 255, 255));
*
* CvInvoke.Imshow("Dilate", img);
* CvInvoke.WaitKey(0);*/
//Filter the contours to only find relevent ones
/* List<Mat> foundOutput = FindandFilterContours(imgClone, img);
*
* for (int i = 0; i < foundOutput.Count; i++)
* {
* CvInvoke.Imshow("Found Output", foundOutput[i]);
* CvInvoke.WaitKey(0);
* }*/
return(img);
}
