void Hu(Image <Gray, byte> img, ref Double[] MomentHu, ref Double[] MomentR)
{
double r1 = 0, r2 = 0, r3 = 0, r4 = 0, r5 = 0, r6 = 0, r7 = 0, r8 = 0, r9 = 0, r10 = 0;
Image <Gray, byte> imgA = new Image <Gray, byte>(img.Size);
imgA = img.ThresholdBinary(new Gray(60), new Gray(255));
Contour <Point> contourA = imgA.FindContours();
MCvMoments momentsA = contourA.GetMoments();
MCvHuMoments HuMomentsA = momentsA.GetHuMoment();
MomentHu[0] = HuMomentsA.hu1;
MomentHu[1] = HuMomentsA.hu2;
MomentHu[2] = HuMomentsA.hu3;
MomentHu[3] = HuMomentsA.hu4;
MomentHu[4] = HuMomentsA.hu5;
MomentHu[5] = HuMomentsA.hu6;
Rmoment(HuMomentsA.hu1, HuMomentsA.hu2, HuMomentsA.hu3, HuMomentsA.hu4, HuMomentsA.hu5, HuMomentsA.hu6, ref r1, ref r2, ref r3, ref r4, ref r5, ref r6, ref r7, ref r8, ref r9, ref r10);
MomentR[0] = r1;
MomentR[1] = r2;
MomentR[2] = r3;
MomentR[3] = r4;
MomentR[4] = r5;
MomentR[5] = r6;
MomentR[6] = r7;
MomentR[7] = r8;
MomentR[8] = r9;
MomentR[9] = r10;
}
/// <summary>
/// 选择轮廓并计算轮廓特征
/// </summary>
private void NumListContours_ValueChanged(object sender, EventArgs e)
{
//先绘制此条轮廓
mArgs.FeatureType = ContourFeatureType.Contour;
mArgs.ContourIndex = (int)NumListContours.Value;
Params_Changed();
//再计算此条轮廓的特征
VectorOfPoint cnt = mContours[mArgs.ContourIndex];
TxtContourArea.Text = CvInvoke.ContourArea(cnt).ToString();
TxtArcLength.Text = CvInvoke.ArcLength(cnt, true).ToString();
MCvMoments moments = CvInvoke.Moments(cnt);
TxtMoments.Text = string.Format("M00: {0}\r\nM01: {1}\r\nM02: {02}\r\nM03: {3}\r\nM10: {4}\r\nM11: {5}\r\nM12: {6}\r\nM20: {7}\r\nM21: {8}\r\nM30: {9}\r\n" +
"Mu02: {10}\r\nMu03: {11}\r\nMu11: {12}\r\nMu12: {13}\r\nMu20: {14}\r\nMu21: {15}\r\nMu30: {16}", moments.M00,
moments.M01, moments.M02, moments.M03, moments.M10, moments.M11, moments.M12, moments.M20, moments.M21, moments.M30,
moments.Mu02, moments.Mu03, moments.Mu11, moments.Mu12, moments.Mu20, moments.Mu21, moments.Mu30);
TxtContourCenter.Text = string.Format("({0},{1})", moments.M10 / moments.M00, moments.M01 / moments.M00);
double[] huMoments = moments.GetHuMoment();
for (int i = 0; i < huMoments.Length; i++)
{
TxtHuMoments.Text += huMoments[i] + "\r\n";
}
}
void Hu(Image <Gray, byte> img, ref Double[] MomentHu)
{
Image <Gray, byte> imgA = new Image <Gray, byte>(img.Size);
imgA = img.ThresholdBinary(new Gray(60), new Gray(255));
Contour <Point> contourA = imgA.FindContours();
MCvMoments momentsA = contourA.GetMoments();
MCvHuMoments HuMomentsA = momentsA.GetHuMoment();
MomentHu[0] = HuMomentsA.hu1;
MomentHu[1] = HuMomentsA.hu2;
MomentHu[2] = HuMomentsA.hu3;
MomentHu[3] = HuMomentsA.hu4;
MomentHu[4] = HuMomentsA.hu5;
MomentHu[5] = HuMomentsA.hu6;
}
private Rectangle LocateROI()
{
int prediction = -1;
VectorOfPoint contourOfInterest = new VectorOfPoint();
int index = 0;
index = ImgProc.LargestContourIndex(_contour);
contourOfInterest = _contour[index];
MCvMoments moment = CvInvoke.Moments(contourOfInterest);
double[] huMoment = moment.GetHuMoment();
prediction = _svm.Compute(huMoment);
//foreach (VectorOfPoint vp in _listOfContours.GetRange(0, 5))
//{
// MCvMoments moment = CvInvoke.Moments(vp);
// double[] huMoment = moment.GetHuMoment();
// prediction = _svm.Compute(huMoment);
// if (prediction == CLASSIFICATION_ARM || prediction == CLASSIFICATION_HAND)
// {
// contourOfInterest = vp;
// break;
// }
//}
if (prediction == CLASSIFICATION_REJECT)
{
return(Rectangle.Empty);
}
else if (prediction == CLASSIFICATION_HAND)
{
//Rectangle rectRotRect = rectRot.MinAreaRect();
//Rectangle init = CvInvoke.MinAreaRect(contoursEval1[largestContourIndexEval1]).MinAreaRect();
//Point final = new Point(rectRotRect.X + init.X, rectRotRect.Y + init.Y);
//return new Rectangle(final, init.Size);
return(CvInvoke.MinAreaRect(contourOfInterest).MinAreaRect());
}
else if (prediction == CLASSIFICATION_ARM)
{
Mat convexityDefect = new Mat();
VectorOfInt hull = new VectorOfInt();
CvInvoke.ConvexHull(contourOfInterest, hull, false, false);
CvInvoke.ConvexityDefects(contourOfInterest, hull, convexityDefect);
RotatedRect rectRot = CvInvoke.MinAreaRect(contourOfInterest);
ModifiedRotatedRect rotRectMod = new ModifiedRotatedRect(rectRot);
int yDel = 0;
double ptLftToRight = Geometry.Distance(rotRectMod.Pul, rotRectMod.Pur);
double ptUpToDown = Geometry.Distance(rotRectMod.Pul, rotRectMod.Pll);
if (!convexityDefect.IsEmpty)
{
Matrix <int> convex = new Matrix <int>(convexityDefect.Rows, convexityDefect.Cols, convexityDefect.NumberOfChannels);
convexityDefect.CopyTo(convex);
List <Point> contourTmp = new List <Point>();
for (int i = 0; i < contourOfInterest.Size; i++)
{
contourTmp.Add(contourOfInterest[i]);
}
List <ConvexDefects> convexDefectList = new List <ConvexDefects>();
for (int i = 0; i < convex.Rows; i++)
{
// do not touch
int startIdx = convex.Data[i, 0];
int endIdx = convex.Data[i, 1];
int pointIdx = convex.Data[i, 2];
Point startPt = contourOfInterest[startIdx];
Point endPt = contourOfInterest[endIdx];
Point defectPt = contourOfInterest[pointIdx];
// do not touch
convexDefectList.Add(new ConvexDefects(startPt, endPt, defectPt));
}
if (ptLftToRight <= ptUpToDown)
{
Point pc1Tmp = convexDefectList[0].DefectPt;
Point pc2Tmp = convexDefectList[1].DefectPt;
Point pc = pc1Tmp.Y > pc2Tmp.Y ? pc1Tmp : pc2Tmp;
Point ptUpLeft = rotRectMod.Pul;
Point ptUpRight = rotRectMod.Pur;
Point ptLowLeft = rotRectMod.Pll;
Point ptLowRight = rotRectMod.Plr;
ModifiedRotatedRect rotRectEval1 = ModifiedRotatedRect.Cut(ptUpLeft, ptUpRight, ptLowLeft, ptLowRight, pc);
ModifiedRotatedRect rotRectEval2 = ModifiedRotatedRect.Cut(ptUpLeft, ptUpRight, ptLowLeft, ptLowRight, pc, true);
Size sizeFrame = ImageInput.Size;
Rectangle rectROIEval1 = rotRectEval1.ToRect(sizeFrame);
Rectangle rectROIEval2 = rotRectEval2.ToRect(sizeFrame);
Mat cloneMat1 = ImageInput.Clone().Mat;
Mat matToBeEval1 = new Mat(cloneMat1, rectROIEval1);
VectorOfVectorOfPoint contoursEval1 = new VectorOfVectorOfPoint();
Mat matHierachyEval1 = new Mat();
CvInvoke.FindContours(matToBeEval1, contoursEval1, matHierachyEval1, Emgu.CV.CvEnum.RetrType.External,
Emgu.CV.CvEnum.ChainApproxMethod.ChainApproxTc89L1);
int largestContourIndexEval1 = ImgProc.LargestContourIndex(contoursEval1);
MCvMoments momentEval1 = CvInvoke.Moments(contoursEval1[largestContourIndexEval1]);
double[] huMomentsEval1 = momentEval1.GetHuMoment();
double[] featureVectorSearch = ScaleValues(huMomentsEval1, 5000.0);
int predictionEval1 = _svm.Compute(featureVectorSearch, MulticlassComputeMethod.Elimination);
//double[] featureVectorHand = ScaleValues(huMomentsEval1.
// .GetRange(0, _svmMachineHand.Inputs).ToArray(), 1000.0);
if (predictionEval1 == CLASSIFICATION_HAND)
{
Rectangle rectRotRect = rectRot.MinAreaRect();
Rectangle init = CvInvoke.MinAreaRect(contoursEval1[largestContourIndexEval1]).MinAreaRect();
Point final = new Point(rectRotRect.X + init.X, rectRotRect.Y + init.Y);
return(new Rectangle(final, init.Size));
}
else
{
return(Rectangle.Empty);
}
}
else
{
return(Rectangle.Empty);
}
}
else
{
return(Rectangle.Empty);
}
}
else
{
return(Rectangle.Empty);
}
}
public void ComputeFeatures(Segment s)
{
//Add the relative size
NamedFeature f = new NamedFeature("RelativeSize");
f.values.Add(s.points.Count() / (double)(imageWidth * imageHeight));
s.features.Add(f);
// Relative centroid
PointF c = NormalizedCentroid(s);
s.features.Add(new NamedFeature("RelativeCentroid", new List <double> {
c.X, c.Y
}));
// One interior point
PointF np = RandomNormalizedInteriorPoint(s);
s.features.Add(new NamedFeature("OneInteriorPoint", new List <double> {
np.X, np.Y
}));
//Radial distance
s.features.Add(new NamedFeature("RadialDistance", new List <double> {
Math.Sqrt(c.X * c.X + c.Y * c.Y)
}));
//Normalized Discrete Compactness http://www.m-hikari.com/imf-password2009/25-28-2009/bribiescaIMF25-28-2009.pdf
//Find the segment id
Point sp = s.points.First();
int sidx = assignments[sp.X, sp.Y];
//count number of perimeter edges
int perimeter = 0;
foreach (Point p in s.points)
{
for (int i = -1; i <= 1; i++)
{
for (int j = -1; j <= 1; j++)
{
if (Math.Abs(i) == Math.Abs(j))
{
continue;
}
if (Util.InBounds(p.X + i, p.Y + j, imageWidth, imageHeight) && assignments[p.X + i, p.Y + j] != sidx)
{
perimeter++;
}
else if (!Util.InBounds(p.X + i, p.Y + j, imageWidth, imageHeight)) //edge pixels should be considered perimeter too
{
perimeter++;
}
}
}
}
int n = s.points.Count();
double CD = (4.0 * n - perimeter) / 2;
double CDmin = n - 1;
double CDmax = (4 * n - 4 * Math.Sqrt(n)) / 2;
double CDN = (CD - CDmin) / Math.Max(1, (CDmax - CDmin));
s.features.Add(new NamedFeature("NormalizedDiscreteCompactness", new List <double> {
CDN
}));
//Add elongation (width/length normalized between 0-square to 1-long http://hal.archives-ouvertes.fr/docs/00/44/60/37/PDF/ARS-Journal-SurveyPatternRecognition.pdf
PointF[] points = s.points.Select <Point, PointF>(p => new PointF(p.X, p.Y)).ToArray <PointF>();
Emgu.CV.Structure.MCvBox2D box = Emgu.CV.PointCollection.MinAreaRect(points);
PointF[] vertices = box.GetVertices();
double elongation = 1 - Math.Min(box.size.Width + 1, box.size.Height + 1) / Math.Max(box.size.Width + 1, box.size.Height + 1);
s.features.Add(new NamedFeature("Elongation", new List <double> {
elongation
}));
//Add Hu shape moments, invariant to translation, scale, and rotation (not sure what each measure refers to intuitively though, or if there is an intuitive analog)
//They may also do badly on noisy data however. See: http://hal.archives-ouvertes.fr/docs/00/44/60/37/PDF/ARS-Journal-SurveyPatternRecognition.pdf (called Invariant Moments)
Bitmap regionBitmap = new Bitmap(imageWidth, imageHeight);
Graphics g = Graphics.FromImage(regionBitmap);
g.FillRectangle(new SolidBrush(Color.Black), 0, 0, imageWidth, imageHeight);
foreach (Point p in s.points)
{
regionBitmap.SetPixel(p.X, p.Y, Color.White);
}
Emgu.CV.Image <Gray, byte> region = new Emgu.CV.Image <Gray, byte>(regionBitmap);
MCvMoments moment = region.GetMoments(true);
MCvHuMoments hu = moment.GetHuMoment();
s.features.Add(new NamedFeature("HuMoments", new List <double> {
hu.hu1, hu.hu2, hu.hu3, hu.hu4, hu.hu5, hu.hu6, hu.hu7
}));
region.Dispose();
regionBitmap.Dispose();
}
