private static Rectangle process(Gray<byte>[,] probabilityMap, Rectangle roi, TermCriteria termCriteria, out CentralMoments centralMoments)
{
Rectangle imageArea = new Rectangle(0, 0, probabilityMap.Width(), probabilityMap.Height());
Rectangle searchWindow = roi;
RawMoments moments = new RawMoments(order: 1);
// Mean shift with fixed number of iterations
int i = 0;
double shift = Byte.MaxValue;
while (termCriteria.ShouldTerminate(i, shift) == false && !searchWindow.IsEmptyArea())
{
// Locate first order moments
moments.Compute(probabilityMap, searchWindow);
int shiftX = (int)(moments.CenterX - searchWindow.Width / 2f);
int shiftY = (int)(moments.CenterY - searchWindow.Height / 2f);
// Shift the mean (centroid)
searchWindow.X += shiftX;
searchWindow.Y += shiftY;
// Keep the search window inside the image
searchWindow.Intersect(imageArea);
shift = System.Math.Abs((double)shiftX) + System.Math.Abs((double)shiftY); //for term criteria only
i++;
}
if (searchWindow.IsEmptyArea() == false)
{
// Locate second order moments and perform final shift
moments.Order = 2;
moments.Compute(probabilityMap, searchWindow);
searchWindow.X += (int)(moments.CenterX - searchWindow.Width / 2f);
searchWindow.Y += (int)(moments.CenterY - searchWindow.Height / 2f);
// Keep the search window inside the image
searchWindow.Intersect(imageArea);
}
centralMoments = new CentralMoments(moments); // moments to be used by camshift
return searchWindow;
}
private unsafe static void computeColor(KernelThread thread, byte* frame, int frameStride, Gray<int>[,] orientationImage, Gray<int>[,] magnitudeSqrImage, int minSqrMagnitude)
{
frame = frame + frameStride * thread.Y + thread.X * 3 /*sizeof(Bgr<byte>)*/;
int maxMagSqr = 0, maxDx = 0, maxDy = 0;
for (int ch = 0; ch < 3; ch++)
{
var srcPtr = frame + ch;
int sumX = 0, sumY = 0;
for (int r = 0; r < 3; r++)
{
var chPtr = srcPtr;
for (int c = 0; c < 3; c++)
{
sumX += *chPtr * Sobel_3x3_X[r, c];
sumY += *chPtr * Sobel_3x3_Y[r, c];
chPtr += 3 * sizeof(byte);
}
srcPtr = (byte*)srcPtr + frameStride;
}
//sumX >>= 3; sumY >>= 3; //divide by 8 (normalize kernel)
var grad = sumX * sumX + sumY * sumY;
if (grad > maxMagSqr)
{
maxMagSqr = grad;
maxDx = sumX;
maxDy = sumY;
}
}
if (maxMagSqr < minSqrMagnitude)
{
//magnitudeSqrImage[thread.Y + kernelRadius, thread.X + kernelRadius] = 0; //redundant
orientationImage[thread.Y + kernelRadius, thread.X + kernelRadius] = FeatureMap.INVALID_ORIENTATION;
}
else
{
magnitudeSqrImage[thread.Y + kernelRadius, thread.X + kernelRadius] = maxMagSqr;
orientationImage[thread.Y + kernelRadius, thread.X + kernelRadius] = MathExtensions.Atan2Aprox(maxDy, maxDx);
}
}
/// <summary>
/// Rectification filter for projective transformation.
/// <para>Accord.NET internal call. Please see: <see cref="Accord.Imaging.Filters.Rectification"/> for details.</para>
/// </summary>
/// <param name="img">Image.</param>
/// <param name="homography">The homography matrix used to map a image passed to the filter to the overlay image.</param>
/// <param name="fillColor">The filling color used to fill blank spaces.</param>
/// <returns>Rectified image.</returns>
public static Gray<byte>[,] Rectification(this Gray<byte>[,] img, double[,] homography, Gray<byte> fillColor)
{
Rectification r = new Rectification(homography);
r.FillColor = fillColor.ToColor();
return img.ApplyBaseTransformationFilter(r);
}
private unsafe static void goodFeaturesToTrack(Gray<float>[,] integralDxx, Gray<float>[,] integralDxy, Gray<float>[,] integralDyy,
int winSize, float minEigValue, Gray<float>[,] strengthImg)
{
minEigValue = System.Math.Max(1E-3f, minEigValue);
int normFactor = winSize * winSize * 255;
int maxCol = integralDxx.Width() - winSize;
int maxRow = integralDxx.Height() - winSize;
for (int row = 0; row < maxRow; row++)
{
for (int col = 0; col < maxCol; col++)
{
var Dxx = integralDxx.GetSum(col, row, winSize, winSize);
var Dxy = integralDxy.GetSum(col, row, winSize, winSize);
var Dyy = integralDyy.GetSum(col, row, winSize, winSize);
var eigenVal = calcMinEigenVal(Dxx, Dxy, Dyy);
eigenVal /= normFactor;
if (eigenVal > minEigValue)
{
strengthImg[winSize / 2 + row, winSize / 2 + col] = eigenVal;
}
}
}
}
/// <summary>
/// Does non-maxima supression for the following gray image. Can be useful for detections filtering (e.g. post-processing output from Harris detector).
/// </summary>
/// <param name="img">Image.</param>
/// <param name="dest">Destination image. Must have the same size as source image.</param>
/// <param name="radius">Non-maxima supression radius (kernel radius).</param>
/// <param name="discardValue">The value will be discarded (0 - for black).</param>
public static void SupressNonMaxima(this Gray <float>[,] img, Gray <float>[,] dest, int radius = 3, int discardValue = 0)
{
using (var uImg = img.Lock())
using (var uDest = dest.Lock())
{
supressNonMaxima_Float(uImg, uDest, radius, discardValue);
}
}
/// <summary>
/// Does non-maxima supression for the following gray image. Can be useful for detections filtering (e.g. post-processing output from Harris detector).
/// </summary>
/// <param name="img">Image.</param>
/// <param name="dest">Destination image. Must have the same size as source image.</param>
/// <param name="radius">Non-maxima supression radius (kernel radius).</param>
/// <param name="discardValue">The value will be discarded (0 - for black).</param>
public static void SupressNonMaxima(this Gray<float>[,] img, Gray<float>[,] dest, int radius = 3, int discardValue = 0)
{
using (var uImg = img.Lock())
using(var uDest = dest.Lock())
{
supressNonMaxima_Float(uImg, uDest, radius, discardValue);
}
}
/// <summary>
/// Does non-maxima supression for the following gray image. Can be useful for detections filtering (e.g. post-processing output from Harris detector).
/// </summary>
/// <param name="img">Image.</param>
/// <param name="radius">Non-maxima supression radius (kernel radius).</param>
/// <param name="discardValue">The value will be discarded (0 - for black).</param>
/// <returns>Processed image.</returns>
public static Gray <float>[,] SupressNonMaxima(this Gray <float>[,] img, int radius = 3, int discardValue = 0)
{
var dest = img.CopyBlank();
SupressNonMaxima(img, dest, radius);
return(dest);
}
public static void Convert(ref Bgr <byte> bgr, ref Gray <byte> gray)
{
int val = ((bgr.R << 1) + //2 * red
(bgr.G << 2) + bgr.G + //5 * green
bgr.B //1 * blue
) >> 3; //divide by 8
gray.Intensity = (byte)val;
}
/// <summary>
/// Converts an image to an bitmap.
/// </summary>
/// <param name="img">Input image.</param>
/// <returns>Bitmap</returns>
public static Bitmap ToBitmap(this Gray <short>[,] img)
{
Bitmap bmp = null;
using (var uImg = img.Lock())
{
bmp = toBitmap(uImg, PixelFormat.Format16bppGrayScale);
}
return(bmp);
}
/// <summary>
/// Converts an image to an bitmap.
/// </summary>
/// <param name="img">Input image.</param>
/// <returns>Bitmap</returns>
public static Bitmap ToBitmap(this Gray <byte>[,] img)
{
Bitmap bmp = null;
using (var uImg = img.Lock())
{
bmp = toBitmap(uImg, PixelFormat.Format8bppIndexed);
}
return(bmp);
}
/// <summary>
/// Creates template from the input image by using provided parameters.
/// </summary>
/// <param name="sourceImage">Input image.</param>
/// <param name="minFeatureStrength">Minimum gradient value for the feature.</param>
/// <param name="maxNumberOfFeatures">Maximum number of features per template. The features will be extracted so that their locations are semi-uniformly spread.</param>
/// <param name="classLabel">Template class label.</param>
public override void Initialize(Gray<byte>[,] sourceImage, int minFeatureStrength, int maxNumberOfFeatures, string classLabel)
{
base.Initialize(sourceImage, minFeatureStrength, maxNumberOfFeatures, classLabel);
this.BinaryMask = sourceImage.Clone(BoundingRect);
if (this.BinaryMask[0, 0].Intensity != 0) //background should be black
BinaryMask = this.BinaryMask.Not();
this.BinaryMask = this.BinaryMask.ThresholdToZero((byte)(255 * 0.75), (byte)255); //if Gauss kernel was applied...
}
/// <summary>
/// Process image looking for corners.
/// </summary>
/// <param name="cornerDetector">Corner detection algorithm instance.</param>
/// <param name="image">Source image to process.</param>
/// <returns>Returns list of found corners (X-Y coordinates).</returns>
public static List<Point> ProcessImage(this ICornersDetector cornerDetector, Gray<byte>[,] image)
{
List<Point> points = null;
using (var uImg = image.Lock())
{
points = cornerDetector.ProcessImage(uImg.AsAForgeImage())
.Select(x => x.ToPoint())
.ToList();
}
return points;
}
/// <summary>
/// Gets specified image portion.
/// If the coordinates are not the rounded, they will be interpolated.
/// </summary>
/// <param name="source">Image.</param>
/// <param name="area">Requested area.</param>
/// <returns>Interpolated image area.</returns>
public static Gray <float>[,] GetRectSubPix(this Gray <float>[,] source, RectangleF area)
{
var destination = new Gray <float> [(int)area.Height, (int)area.Width];
using (var srcImg = source.Lock())
using (var dstImg = destination.Lock())
{
getRectSubPix_Float(srcImg, area.Location, dstImg);
}
return(destination);
}
/// <summary>
/// Gray-Level Difference Method (GLDM).
/// <para>Computes an gray-level histogram of difference values between adjacent pixels in an image.</para>
/// <para>Accord.NET internal call. Please see: <see cref="Accord.Imaging.GrayLevelDifferenceMethod">Gray-Level Difference Method</see> for details.</para>
/// </summary>
/// <param name="image">The source image.</param>
/// <param name="autoGray">Whether the maximum value of gray should be automatically computed from the image. </param>
/// <param name="degree">The direction at which the co-occurrence should be found.</param>
/// <returns>An histogram containing co-occurrences for every gray level in <paramref name="image"/>.</returns>
public static int[] GrayLevelDifferenceMethod(this Gray <byte>[,] image, CooccurrenceDegree degree, bool autoGray = true)
{
GrayLevelDifferenceMethod gldm = new GrayLevelDifferenceMethod(degree, autoGray);
int[] hist;
using (var uImg = image.Lock())
{
hist = gldm.Compute(uImg.AsAForgeImage());
}
return(hist);
}
/// <summary>
/// Maximum cross-correlation feature point matching algorithm.
/// </summary>
/// <param name="image1">First image.</param>
/// <param name="image2">Second image.</param>
/// <param name="points1">Points from the first image.</param>
/// <param name="points2">Points from the second image.</param>
/// <param name="windowSize">The size of the correlation window.</param>
/// <param name="maxDistance">The maximum distance to consider points as correlated.</param>
/// <returns>Matched point-pairs.</returns>
public static Point[][] Match(Gray<byte>[,] image1, Gray<byte>[,] image2, Point[] points1, Point[] points2, int windowSize, int maxDistance)
{
Point[][] matches = null;
using (var uImg1 = image1.Lock())
using(var uImg2 = image2.Lock())
{
var correlationMatching = new CorrelationMatching(windowSize, maxDistance, uImg1.AsBitmap(), uImg2.AsBitmap());
matches = correlationMatching.Match(points1.ToPoints(), points2.ToPoints()).ToPoints();
}
return matches;
}
/// <summary>
/// Features from Accelerated Segment Test (FAST) corners detector.
/// <para>Accord.NET internal call. Please see: <see cref="Accord.Imaging.FastCornersDetector"/> for details.</para>
/// </summary>
/// <param name="im">Image.</param>
/// <param name="threshold">The suppression threshold. Decreasing this value increases the number of points detected by the algorithm.</param>
/// <returns>Interest point locations.</returns>
public static List <IntPoint> CornerFeaturesDetector(this Gray <byte>[,] im, int threshold = 20)
{
FastCornersDetector fast = new FastCornersDetector(threshold);
List <IntPoint> points;
using (var uImg = im.Lock())
{
points = fast.ProcessImage(uImg.AsAForgeImage());
}
return(points);
}
/// <summary>
/// Back-projects (creates probability map) from histogram values.
/// </summary>
/// <param name="srcs">Image channels.</param>
/// <returns>Back-projection image (probability image) </returns>
public Gray<byte>[,] BackProject(Gray<byte>[][,] srcs)
{
var destImg = srcs.First().CopyBlank();
using (var uDestImg = destImg.Lock())
{
var uChannels = srcs.Select(x => x.Lock()).ToArray();
backProjectByte(this, uChannels, uDestImg);
uChannels.ForEach(x => x.Dispose());
}
return destImg;
}
/// <summary>
/// Harris Corners Detector.
/// <para>Accord.NET internal call. Please see: <see cref="Accord.Imaging.HarrisCornersDetector"/> for details.</para>
/// </summary>
/// <param name="im">Image.</param>
/// <param name="measure">Corners measures.</param>
/// <param name="threshold">Harris threshold.</param>
/// <param name="sigma">Gaussian smoothing sigma.</param>
/// <param name="suppression">Non-maximum suppression window radius.</param>
/// <returns>Interest point locations.</returns>
public static List <IntPoint> HarrisCorners <TDepth>(this Gray <byte>[,] im, HarrisCornerMeasure measure = HarrisCornerMeasure.Harris, float threshold = 20000f, double sigma = 1.2, int suppression = 3)
{
HarrisCornersDetector harris = new HarrisCornersDetector(measure, threshold, sigma, suppression);
List <IntPoint> points;
using (var uImg = im.Lock())
{
points = harris.ProcessImage(uImg.AsAForgeImage());
}
return(points);
}
/// <summary>
/// Extracts the contour from a single object in a grayscale image. (uses Accord built-in function)
/// </summary>
/// <param name="im">Image.</param>
/// <param name="minGradientStrength">The pixel value threshold above which a pixel
/// is considered black (belonging to the object). Default is zero.</param>
public static List <Point> FindContour(this Gray <byte>[,] im, byte minGradientStrength = 0)
{
BorderFollowing bf = new BorderFollowing(minGradientStrength);
List <Point> points;
using (var uImg = im.Lock())
{
points = bf.FindContour(uImg.AsAForgeImage());
}
return(points);
}
/// <summary>
/// Find non-zero locations in the image.
/// </summary>
/// <param name="img">Image.</param>
/// <param name="values">Found non-zero values at the returned positions.</param>
/// <returns>List of found non-zero locations.</returns>
public static List <Point> FindNonZero(this Gray <float>[,] img, out IList <float> values)
{
List <Point> locationsPatch;
IList valuesPatch;
using (var uImg = img.Lock())
{
findNonZero_Float(uImg, out locationsPatch, out valuesPatch);
}
values = valuesPatch as IList <float>;
return(locationsPatch);
}
/// <summary>
/// Back-projects (creates probability map) from histogram values.
/// </summary>
/// <param name="srcs">Image channels.</param>
/// <returns>Back-projection image (probability image) </returns>
public Gray <byte>[,] BackProject(Gray <byte>[][,] srcs)
{
var destImg = srcs.First().CopyBlank();
using (var uDestImg = destImg.Lock())
{
var uChannels = srcs.Select(x => x.Lock()).ToArray();
backProjectByte(this, uChannels, uDestImg);
uChannels.ForEach(x => x.Dispose());
}
return(destImg);
}
/// <summary>
/// Maximum cross-correlation feature point matching algorithm.
/// </summary>
/// <param name="image1">First image.</param>
/// <param name="image2">Second image.</param>
/// <param name="points1">Points from the first image.</param>
/// <param name="points2">Points from the second image.</param>
/// <param name="windowSize">The size of the correlation window.</param>
/// <param name="maxDistance">The maximum distance to consider points as correlated.</param>
/// <returns>Matched point-pairs.</returns>
public static Point[][] Match(Gray <byte>[,] image1, Gray <byte>[,] image2, Point[] points1, Point[] points2, int windowSize, int maxDistance)
{
Point[][] matches = null;
using (var uImg1 = image1.Lock())
using (var uImg2 = image2.Lock())
{
var correlationMatching = new CorrelationMatching(windowSize, maxDistance, uImg1.AsBitmap(), uImg2.AsBitmap());
matches = correlationMatching.Match(points1.ToPoints(), points2.ToPoints()).ToPoints();
}
return(matches);
}
/// <summary>
/// The blending filter is able to blend two images using a homography matrix.
/// A linear alpha gradient is used to smooth out differences between the two
/// images, effectively blending them in two images. The gradient is computed
/// considering the distance between the centers of the two images.
/// </summary>
/// <param name="im">Image.</param>
/// <param name="overlayIm">The overlay image (also called the anchor).</param>
/// <param name="homography">Homography matrix used to map a image passed to
/// the filter to the overlay image specified at filter creation.</param>
/// <param name="fillColor">The filling color used to fill blank spaces. The filling color will only be visible after the image is converted
/// to 24bpp. The alpha channel will be used internally by the filter.</param>
/// <param name="gradient">A value indicating whether to blend using a linear
/// gradient or just superimpose the two images with equal weights.</param>
/// <param name="alphaOnly">A value indicating whether only the alpha channel
/// should be blended. This can be used together with a transparency
/// mask to selectively blend only portions of the image.</param>
/// <returns>Blended image.</returns>
public static Bgra <byte>[,] Blend(this Gray <byte>[,] im, Gray <byte>[,] overlayIm, MatrixH homography, Bgra <byte> fillColor, bool gradient = true, bool alphaOnly = false)
{
Bgra <byte>[,] resultImage = null;
using (var uOverlayIm = overlayIm.Lock())
{
Blend blend = new Blend(homography, uOverlayIm.AsBitmap());
blend.AlphaOnly = alphaOnly;
blend.Gradient = gradient;
blend.FillColor = fillColor.ToColor();
resultImage = im.ApplyBaseTransformationFilter <Gray <byte>, Bgra <byte> >(blend);
}
return(resultImage);
}
/// <summary>
/// The blending filter is able to blend two images using a homography matrix.
/// A linear alpha gradient is used to smooth out differences between the two
/// images, effectively blending them in two images. The gradient is computed
/// considering the distance between the centers of the two images.
/// </summary>
/// <param name="im">Image.</param>
/// <param name="overlayIm">The overlay image (also called the anchor).</param>
/// <param name="homography">Homography matrix used to map a image passed to
/// the filter to the overlay image specified at filter creation.</param>
/// <param name="fillColor">The filling color used to fill blank spaces. The filling color will only be visible after the image is converted
/// to 24bpp. The alpha channel will be used internally by the filter.</param>
/// <param name="gradient">A value indicating whether to blend using a linear
/// gradient or just superimpose the two images with equal weights.</param>
/// <param name="alphaOnly">A value indicating whether only the alpha channel
/// should be blended. This can be used together with a transparency
/// mask to selectively blend only portions of the image.</param>
/// <returns>Blended image.</returns>
public static Bgra<byte>[,] Blend(this Gray<byte>[,] im, Gray<byte>[,] overlayIm, MatrixH homography, Bgra<byte> fillColor, bool gradient = true, bool alphaOnly = false)
{
Bgra<byte>[,] resultImage = null;
using (var uOverlayIm = overlayIm.Lock())
{
Blend blend = new Blend(homography, uOverlayIm.AsBitmap());
blend.AlphaOnly = alphaOnly;
blend.Gradient = gradient;
blend.FillColor = fillColor.ToColor();
resultImage = im.ApplyBaseTransformationFilter<Gray<byte>, Bgra<byte>>(blend);
}
return resultImage;
}
/// <summary>
/// Computes the Bag of Words model.
/// </summary>
/// <typeparam name="TPoint">
/// The <see cref="Accord.Imaging.IFeaturePoint{TFeature}"/> type to be used with this class,
/// such as <see cref="Accord.Imaging.SpeededUpRobustFeaturePoint"/>.
/// </typeparam>
/// <typeparam name="TFeature">
/// The feature type of the <typeparamref name="TPoint"/>, such
/// as <see cref="T:double[]"/>.
/// </typeparam>
/// <param name="bow">Bag of Visual Words.</param>
/// <param name="images">The set of images to initialize the model.</param>
/// <param name="threshold">Convergence rate for the k-means algorithm. Default is 1e-5.</param>
/// <returns>The list of feature points detected in all images.</returns>
public static List <TPoint>[] Compute <TPoint, TFeature>(this BagOfVisualWords <TPoint, TFeature> bow,
Gray <byte>[][,] images, double threshold = 1e-5)
where TPoint : IFeatureDescriptor <TFeature>
{
var uImages = images.Select(x => x.Lock());
var featurePoints = bow.Compute
(
uImages.Select(x => x.AsBitmap()).ToArray(),
threshold
);
uImages.ForEach(x => x.Dispose());
return(featurePoints);
}
/// <summary>
/// Extracts the specified image channels.
/// </summary>
/// <typeparam name="TSrcColor">Source color type.</typeparam>
/// <typeparam name="TDepth">Channel depth type.</typeparam>
/// <param name="image">Image.</param>
/// <param name="area">Working area.</param>
/// <param name="channelIndices">Channel indicies to extract. If null, all channels are extracted.</param>
/// <returns>Channel collection.</returns>
public static unsafe Gray <TDepth>[][,] SplitChannels <TSrcColor, TDepth>(this TSrcColor[,] image, Rectangle area, params int[] channelIndices)
where TSrcColor : struct, IColor <TDepth>
where TDepth : struct
{
if (channelIndices == null || channelIndices.Length == 0)
{
channelIndices = Enumerable.Range(0, ColorInfo.GetInfo <TSrcColor>().ChannelCount).ToArray();
}
var channels = new Gray <TDepth> [channelIndices.Length][, ];
for (int i = 0; i < channelIndices.Length; i++)
{
channels[i] = GetChannel <TSrcColor, TDepth>(image, area, channelIndices[i]);
}
return(channels);
}
private static Rectangle process(Gray <byte>[,] probabilityMap, Rectangle roi, TermCriteria termCriteria, out CentralMoments centralMoments)
{
Rectangle imageArea = new Rectangle(0, 0, probabilityMap.Width(), probabilityMap.Height());
Rectangle searchWindow = roi;
RawMoments moments = new RawMoments(order: 1);
// Mean shift with fixed number of iterations
int i = 0;
double shift = Byte.MaxValue;
while (termCriteria.ShouldTerminate(i, shift) == false && !searchWindow.IsEmptyArea())
{
// Locate first order moments
moments.Compute(probabilityMap, searchWindow);
int shiftX = (int)(moments.CenterX - searchWindow.Width / 2f);
int shiftY = (int)(moments.CenterY - searchWindow.Height / 2f);
// Shift the mean (centroid)
searchWindow.X += shiftX;
searchWindow.Y += shiftY;
// Keep the search window inside the image
searchWindow.Intersect(imageArea);
shift = System.Math.Abs((double)shiftX) + System.Math.Abs((double)shiftY); //for term criteria only
i++;
}
if (searchWindow.IsEmptyArea() == false)
{
// Locate second order moments and perform final shift
moments.Order = 2;
moments.Compute(probabilityMap, searchWindow);
searchWindow.X += (int)(moments.CenterX - searchWindow.Width / 2f);
searchWindow.Y += (int)(moments.CenterY - searchWindow.Height / 2f);
// Keep the search window inside the image
searchWindow.Intersect(imageArea);
}
centralMoments = new CentralMoments(moments); // moments to be used by camshift
return(searchWindow);
}
private static void createRange(Dictionary<ModelParams, ITemplate> dict, Gray<byte>[,] templateImg,
int templateIdx, IEnumerable<int> scaleRange, IEnumerable<int> rotationRange, string label = "")
{
var contour = findContour(templateImg);
var pts = contour.GetEqualyDistributedPoints(N_SAMPLE_POINTS, treatAsClosed: false);
pts = pts.Normalize();
foreach(var s in scaleRange)
{
foreach(var r in rotationRange)
{
var template = create(pts, s, r, label);
var mParams = new ModelParams(templateIdx, (short)s, (short)r);
dict.Add(mParams, template);
}
}
}
/// <summary>
/// Extracts a single image channel.
/// </summary>
/// <typeparam name="TSrcColor">Source color type.</typeparam>
/// <typeparam name="TDepth">Channel depth type.</typeparam>
/// <param name="image">Image.</param>
/// <param name="area">Working area.</param>
/// <param name="channelIndex">Channel index.</param>
/// <returns>Extracted channel.</returns>
public static unsafe Gray <TDepth>[,] GetChannel <TSrcColor, TDepth>(this TSrcColor[,] image, Rectangle area, int channelIndex)
where TSrcColor : struct, IColor <TDepth>
where TDepth : struct
{
int width = area.Width;
int height = area.Height;
var dest = new Gray <TDepth> [area.Height, area.Width];
using (var lockedImage = image.Lock())
using (var dstImg = dest.Lock())
{
var srcImg = lockedImage.GetSubRect(area);
int channelSize = srcImg.ColorInfo.ChannelSize;
int colorSize = srcImg.ColorInfo.Size;
byte *srcPtr = (byte *)srcImg.ImageData + channelIndex * srcImg.ColorInfo.ChannelSize;
byte *dstPtr = (byte *)dstImg.ImageData;
for (int row = 0; row < height; row++)
{
byte *srcColPtr = srcPtr;
byte *dstColPtr = dstPtr;
for (int col = 0; col < width; col++)
{
/********** copy channel byte-per-byte ************/
for (int partIdx = 0; partIdx < channelSize; partIdx++)
{
dstColPtr[partIdx] = srcColPtr[partIdx];
}
srcColPtr += colorSize; //move to the next column
dstColPtr += channelSize;
/********** copy channel byte-per-byte ************/
}
srcPtr += srcImg.Stride;
dstPtr += dstImg.Stride;
}
}
return(dest);
}
/// <summary>
/// Calculates histogram.
/// </summary>
/// <param name="channels">Image channels.</param>
/// <param name="accumulate">Accumulate or erase histogram before.</param>
/// <param name="mask">Mask for image color locations.</param>
/// <param name="maskOffset">The location offset for the mask. The mask area will be [offsetX, offsetY, channelWidth, channelHeight].</param>
public void Calculate(Gray<byte>[][,] channels, bool accumulate, Gray<byte>[,] mask, Point maskOffset)
{
if (!accumulate)
Array.Clear(histogram, 0, this.NumberOfElements);
if (mask == null)
{
mask = new Gray<byte>[channels[0].Width(), channels[0].Height()];
mask.SetValue<Gray<byte>>(Byte.MaxValue);
}
var maskArea = new Rectangle(maskOffset, channels.First().Size());
using (var uMask = mask.Lock(maskArea))
{
var uChannels = channels.Select(x => x.Lock()).ToArray();
calculateHistByte(this, uChannels, uMask);
uChannels.ForEach(x => x.Dispose());
}
}
/// <summary>
/// Computes gradient orientations from the color image. Orientation from the channel which has the maximum gradient magnitude is taken as the orientation for a location.
/// </summary>
/// <param name="frame">Image.</param>
/// <param name="magnitudeSqrImage">Squared magnitude image.</param>
/// <param name="minValidMagnitude">Minimal valid magnitude.</param>
/// <returns>Orientation image (angles are in degrees).</returns>
public static unsafe Gray<int>[,] Compute(Bgr<byte>[,] frame, out Gray<int>[,] magnitudeSqrImage, int minValidMagnitude)
{
var minSqrMagnitude = minValidMagnitude * minValidMagnitude;
var orientationImage = new Gray<int>[frame.Height(), frame.Width()];
var _magnitudeSqrImage = orientationImage.CopyBlank();
using (var uFrame = frame.Lock())
{
ParallelLauncher.Launch(thread =>
{
computeColor(thread, (byte*)uFrame.ImageData, uFrame.Stride, orientationImage, _magnitudeSqrImage, minSqrMagnitude);
},
frame.Width() - 2 * kernelRadius, frame.Height() - 2 * kernelRadius);
}
magnitudeSqrImage = _magnitudeSqrImage;
return orientationImage;
}
private static Gray <float>[,] convolve(Gray <float>[,] image, IList <float[, ]> kernels, ConvolutionBorder options)
{
int biggestKernelWidth, biggestKernelHeight;
getTheBiggestSize(kernels, out biggestKernelWidth, out biggestKernelHeight);
int fillX, fillY;
var paddedIm = prepareImage(image, biggestKernelWidth, biggestKernelHeight, options, out fillX, out fillY);
var convolvedIm = paddedIm;
foreach (var kernel in kernels)
{
var preparedKernel = prepareKernel(kernel, convolvedIm.Size());
convolvedIm = convolvedIm.MulComplex(preparedKernel, inPlace: false);
}
return(getConvolutionResult(convolvedIm, fillX, fillY, image.Size()));
}
private static IList<PointF> findContour(Gray<byte>[,] templateImg)
{
var contour = templateImg.FindContour(minGradientStrength: 150).Select(x => (PointF)x).ToList();
/*********** cut bottom border and shift contour beginning to the first non-border point ***************/
int firstIdx = -1;
int lastIdx = -1;
for (int i = 0; i < contour.Count; i++)
{
if (contour[i].Y == (templateImg.Height() - 1))
{
if (firstIdx == -1) firstIdx = i;
lastIdx = i;
}
}
//return contour;
return new CircularList<PointF>(contour).GetRange(lastIdx, contour.Count - (lastIdx - firstIdx + 1));
}
private void processImage(Bgr<byte>[,] frame, out Gray<byte>[,] probabilityMap, out Rectangle prevSearchArea, out Box2D foundBox)
{
prevSearchArea = searchArea;
//convert to HSV
var hsvImg = frame.ToHsv();
//back-project ratio hist => create probability map
probabilityMap = ratioHist.BackProject(hsvImg.SplitChannels<Hsv<byte>, byte>(0, 1)); //or new Image<Gray<byte>>[]{ hsvImg[0], hsvImg[1]...}
//user constraints...
Gray<byte>[,] mask = hsvImg.InRange(new Hsv<byte>(0, 0, (byte)minV), new Hsv<byte>(0, 0, (byte)maxV), Byte.MaxValue, 2);
probabilityMap.AndByte(mask, inPlace:true);
//run Camshift algorithm to find new object position, size and angle
foundBox = Camshift.Process(probabilityMap, searchArea);
var foundArea = Rectangle.Round(foundBox.GetMinArea());
searchArea = foundArea.Inflate(0.05, 0.05, frame.Size()); //inflate found area for search (X factor)...
if (searchArea.IsEmpty) isROISelected = false; //reset tracking
}
/// <summary>
/// Camshift algorithm
/// </summary>
/// <param name="probabilityMap">Probability map [0-255].</param>
/// <param name="roi">Initial Search area</param>
/// <param name="termCriteria">Mean shift termination criteria (PLEASE DO NOT REMOVE (but you can move it) THIS CLASS; PLEASE!!!)</param>
/// <param name="centralMoments">Calculated central moments (up to order 2).</param>
/// <returns>Object position, size and angle packed into a structure.</returns>
public static Box2D Process(Gray <byte>[,] probabilityMap, Rectangle roi, TermCriteria termCriteria, out CentralMoments centralMoments)
{
// Compute mean shift
Rectangle objArea = Meanshift.Process(probabilityMap, roi, termCriteria, out centralMoments);
//fit ellipse
Ellipse ellipse = centralMoments.GetEllipse();
ellipse.Center = objArea.Center();
//return empty structure is the object is lost
var sz = ellipse.Size;
if (Single.IsNaN(sz.Width) || Single.IsNaN(sz.Height) ||
sz.Width < 1 || sz.Height < 1)
{
return(Box2D.Empty);
}
return((Box2D)ellipse);
}
/// <summary>
/// Calculates histogram.
/// </summary>
/// <param name="channels">Image channels.</param>
/// <param name="accumulate">Accumulate or erase histogram before.</param>
/// <param name="mask">Mask for image color locations.</param>
/// <param name="maskOffset">The location offset for the mask. The mask area will be [offsetX, offsetY, channelWidth, channelHeight].</param>
public void Calculate(Gray <byte>[][,] channels, bool accumulate, Gray <byte>[,] mask, Point maskOffset)
{
if (!accumulate)
{
Array.Clear(histogram, 0, this.NumberOfElements);
}
if (mask == null)
{
mask = new Gray <byte> [channels[0].Width(), channels[0].Height()];
mask.SetValue <Gray <byte> >(Byte.MaxValue);
}
var maskArea = new Rectangle(maskOffset, channels.First().Size());
using (var uMask = mask.Lock(maskArea))
{
var uChannels = channels.Select(x => x.Lock()).ToArray();
calculateHistByte(this, uChannels, uMask);
uChannels.ForEach(x => x.Dispose());
}
}
public void TestLKFlow()
{
var im1 = new Gray<float>[480, 640];
im1.SetValue<Gray<float>>(System.Byte.MaxValue, new Accord.Extensions.Rectangle(272, 82, 116, 64));
var im2 = new Gray<float>[480, 640];
im2.SetValue<Gray<float>>(System.Byte.MaxValue, new Accord.Extensions.Rectangle(277, 83, 116, 64));
var pts = new List<PointF>();
pts.Add(new PointF(272, 82)); //-> 277,83
PointF[] currFeatures;
KLTFeatureStatus[] featureStatus;
/*LKOpticalFlow<Gray<float>>.EstimateFlow(lkStorage, pts.ToArray(),
out currFeatures, out featureStatus);*/
PyrLKOpticalFlow<Gray<float>>.EstimateFlow(im1, im2, pts.ToArray(),
out currFeatures, out featureStatus);
Assert.IsTrue(featureStatus[0] == KLTFeatureStatus.Success);
Assert.IsTrue(Math.Round(currFeatures[0].X) == 277 && Math.Round(currFeatures[0].Y) == 83);
}
private void initTracking(Bgr <byte>[,] frame)
{
//get hue channel from search area
var hsvImg = frame.ToHsv();
//user constraints...
Gray <byte>[,] mask = hsvImg.InRange(new Hsv <byte>(0, 0, (byte)minV), new Hsv <byte>(0, 0, (byte)maxV), Byte.MaxValue, 2);
originalObjHist.Calculate(hsvImg.SplitChannels <Hsv <byte>, byte>(roi, 0, 1), !false, mask, roi.Location);
originalObjHist.Scale((float)1 / roi.Area());
//originalObjHist.Normalize(Byte.MaxValue);
var backgroundArea = roi.Inflate(1.5, 1.5, frame.Size());
backgroundHist.Calculate(hsvImg.SplitChannels <Hsv <byte>, byte>(backgroundArea, 0, 1), !false, mask, backgroundArea.Location);
backgroundHist.Scale((float)1 / backgroundArea.Area());
//backgroundHist.Normalize(Byte.MaxValue);
//how good originalObjHist and objHist match (suppresses possible selected background)
ratioHist = originalObjHist.CreateRatioHistogram(backgroundHist, Byte.MaxValue, 10);
searchArea = roi;
roi = Rectangle.Empty;
}
private void processImage(Bgr <byte>[,] frame, out Gray <byte>[,] probabilityMap, out Rectangle prevSearchArea, out Box2D foundBox)
{
prevSearchArea = searchArea;
//convert to HSV
var hsvImg = frame.ToHsv();
//back-project ratio hist => create probability map
probabilityMap = ratioHist.BackProject(hsvImg.SplitChannels <Hsv <byte>, byte>(0, 1)); //or new Image<Gray<byte>>[]{ hsvImg[0], hsvImg[1]...}
//user constraints...
Gray <byte>[,] mask = hsvImg.InRange(new Hsv <byte>(0, 0, (byte)minV), new Hsv <byte>(0, 0, (byte)maxV), Byte.MaxValue, 2);
probabilityMap.AndByte(mask, inPlace: true);
//run Camshift algorithm to find new object position, size and angle
foundBox = Camshift.Process(probabilityMap, searchArea);
var foundArea = Rectangle.Round(foundBox.GetMinArea());
searchArea = foundArea.Inflate(0.05, 0.05, frame.Size()); //inflate found area for search (X factor)...
if (searchArea.IsEmpty)
{
isROISelected = false; //reset tracking
}
}
/// <summary>
/// Resizes an image using specified interpolation mode.
/// </summary>
/// <param name="img">Input image.</param>
/// <param name="scale">Non-negative image size scale factor. If 1 the new size will be equal.</param>
/// <param name="mode">Interpolation mode.</param>
/// <returns>Resized image.</returns>
public static Gray <byte>[,] Resize(this Gray <byte>[,] img, float scale, InterpolationMode mode)
{
return(ResizeExtensionsBase.Resize <Gray <byte> >(img, scale, mode));
}
/// <summary>
/// Meanshift algorithm
/// </summary>
/// <param name="probabilityMap">Probability map [0-1].</param>
/// <param name="roi">Initial search area</param>
/// <param name="termCriteria">Mean shift termination criteria.</param>
/// <returns>Object area.</returns>
public static Rectangle Process(Gray<byte>[,] probabilityMap, Rectangle roi, TermCriteria termCriteria)
{
CentralMoments centralMoments;
return process(probabilityMap, roi, termCriteria, out centralMoments);
}
/// <summary>
/// Calculates phase using Atan2 (secondImage / firstImage).
/// </summary>
/// <param name="imageX">First image.</param>
/// <param name="imageY">Second image.</param>
/// <returns>Phase.</returns>
public static Gray <double>[,] Phase(this Gray <double>[,] imageX, Gray <double>[,] imageY)
{
return(imageX.Calculate(imageY, phase_Double, inPlace: false));
}
/// <summary>
/// Calculates phase using Atan2 (secondImage / firstImage).
/// </summary>
/// <param name="imageX">First image.</param>
/// <param name="imageY">Second image.</param>
/// <returns>Phase.</returns>
public static Gray <float>[,] Phase(this Gray <float>[,] imageX, Gray <float>[,] imageY)
{
return(imageX.Calculate(imageY, phase_Float, inPlace: false));
}
/// <summary>
/// Calculates phase using Atan2 (secondImage / firstImage).
/// </summary>
/// <param name="imageX">First image.</param>
/// <param name="imageY">Second image.</param>
/// <returns>Phase.</returns>
public static Gray<double>[,] Phase(this Gray<double>[,] imageX, Gray<double>[,] imageY)
{
return imageX.Calculate(imageY, phase_Double, inPlace: false);
}
/// <summary>
/// Camshift algorithm
/// </summary>
/// <param name="probabilityMap">Probability map [0-1].</param>
/// <param name="roi">Initial Search area</param>
/// <returns>Object position, size and angle packed into a structure.</returns>
public static Box2D Process(Gray <byte>[,] probabilityMap, Rectangle roi)
{
CentralMoments centralMoments;
return(Process(probabilityMap, roi, Meanshift.DEFAULT_TERM, out centralMoments));
}
private static unsafe void computeGray(KernelThread thread, byte* frame, int frameStride, Gray<int>[,] orientationImage, Gray<int>[,] magnitudeSqrImage, int minSqrMagnitude)
{
frame = frame + frameStride * thread.Y + thread.X;
var srcPtr = frame;
int sumX = 0, sumY = 0;
for (int r = 0; r < 3; r++)
{
for (int c = 0; c < 3; c++)
{
sumX += srcPtr[c] * Sobel_3x3_X[r, c];
sumY += srcPtr[c] * Sobel_3x3_Y[r, c];
}
srcPtr = (byte*)srcPtr + frameStride;
}
//sumX >>= 3; sumY >>= 3; //divide by 8 (normalize kernel) //without this
var grad = sumX * sumX + sumY * sumY;
if (grad < minSqrMagnitude)
{
//magnitudeSqrImage[thread.Y + kernelRadius, thread.X + kernelRadius] = 0; //redundant
orientationImage[thread.Y + kernelRadius, thread.X + kernelRadius] = FeatureMap.INVALID_ORIENTATION;
}
else
{
magnitudeSqrImage[thread.Y + kernelRadius, thread.X + kernelRadius] = grad;
orientationImage[thread.Y + kernelRadius, thread.X + kernelRadius] = MathExtensions.Atan2Aprox(sumY, sumX);
}
}
/// <summary>
/// The blending filter is able to blend two images using a homography matrix.
/// A linear alpha gradient is used to smooth out differences between the two
/// images, effectively blending them in two images. The gradient is computed
/// considering the distance between the centers of the two images.
/// <para>Homography matrix is set to identity.</para>
/// <para>Fill color is set to black with alpha set to 0 (all zeros).</para>
/// </summary>
/// <param name="im">Image.</param>
/// <param name="overlayIm">The overlay image (also called the anchor).</param>
/// <param name="gradient">A value indicating whether to blend using a linear
/// gradient or just superimpose the two images with equal weights.</param>
/// <param name="alphaOnly">A value indicating whether only the alpha channel
/// should be blended. This can be used together with a transparency
/// mask to selectively blend only portions of the image.</param>
/// <returns>Blended image.</returns>
public static Bgra <byte>[,] Blend(this Gray <byte>[,] im, Gray <byte>[,] overlayIm, bool gradient = true, bool alphaOnly = false)
{
return(Blend(im, overlayIm, new MatrixH(Matrix.Identity(3)), new Bgra <byte>(), gradient, alphaOnly));
}
private void trackCamshift(Bgr<byte>[,] frame, Rectangle searchArea, out Gray<byte>[,] probabilityMap, out Box2D foundBox)
{
const int PROBABILITY_MIN_VAL = (int)(0.3f * Byte.MaxValue);
//convert to HSV
var hsvImg = frame.ToHsv();
//back-project ratio hist => create probability map
probabilityMap = ratioHist.BackProject(hsvImg.SplitChannels<Hsv<byte>, byte>(0, 1)); //or new Image<Gray<byte>>[]{ hsvImg[0], hsvImg[1]...}
//user constraints...
Gray<byte>[,] mask = hsvImg.InRange(new Hsv<byte>(0, 0, (byte)minV), new Hsv<byte>(0, 0, (byte)maxV), Byte.MaxValue, 2);
probabilityMap.AndByte(mask, inPlace: true);
//run Camshift algorithm to find new object position, size and angle
CentralMoments centralMoments;
foundBox = Camshift.Process(probabilityMap, searchArea, Meanshift.DEFAULT_TERM, out centralMoments);
//stopping conditions
float avgIntensity = centralMoments.Mu00 / (foundBox.Size.Area() + Single.Epsilon);
if (avgIntensity < PROBABILITY_MIN_VAL || foundBox.Size.IsEmpty || foundBox.Size.Height < 12)
{
foundBox = Box2D.Empty; //invalid box
}
}
private void trackOneStep(Bgr<byte>[,] frame, out Gray<byte>[,] probabilityMap, out Box2D foundBox)
{
const float SEARCH_AREA_INFLATE_FACTOR = 0.05f;
/**************************** KALMAN predict **************************/
kalman.Predict();
searchArea = createRect(kalman.State.Position, searchArea.Size, frame.Size());
/**************************** KALMAN predict **************************/
trackCamshift(frame, searchArea, out probabilityMap, out foundBox);
if (!foundBox.IsEmpty)
{
/**************************** KALMAN correct **************************/
kalman.Correct(new PointF(foundBox.Center.X, foundBox.Center.Y)); //correct predicted state by measurement
/**************************** KALMAN correct **************************/
var foundArea = Rectangle.Round(foundBox.GetMinArea());
searchArea = foundArea.Inflate(SEARCH_AREA_INFLATE_FACTOR, SEARCH_AREA_INFLATE_FACTOR, frame.Size()); //inflate found area for search (X factor)...
nonVisibleCount = 0;
}
else
{
nonVisibleCount++;
if (nonVisibleCount == 1) //for the first time
{
searchArea = searchArea.Inflate(-SEARCH_AREA_INFLATE_FACTOR * 1.5, -SEARCH_AREA_INFLATE_FACTOR * 1.5, frame.Size()); //shrink (hysteresis)
}
searchArea = createRect(kalman.State.Position, searchArea.Size, frame.Size());
}
if (nonVisibleCount > 100) //if not visible for a longer time => reset tracking
{
nonVisibleCount = 0;
isROISelected = false;
}
}
private static List<Feature> ExtractTemplate(Gray<byte>[,] orientationImage, int maxNumOfFeatures, Func<Feature, int> featureImportanceFunc)
{
List<Feature> candidates = new List<Feature>();
using (var uOrientationImage = orientationImage.Lock())
{
byte* orientImgPtr = (byte*)uOrientationImage.ImageData;
int orientImgStride = uOrientationImage.Stride;
int imgWidth = uOrientationImage.Width;
int imgHeight = uOrientationImage.Height;
for (int row = 0; row < imgHeight; row++)
{
for (int col = 0; col < imgWidth; col++)
{
if (orientImgPtr[col] == 0) //quantized orientations are: [1,2,4,8,...,128];
continue;
var candidate = new Feature(x: col, y: row, angleBinaryRepresentation: orientImgPtr[col]);
candidates.Add(candidate);
}
orientImgPtr += orientImgStride;
}
}
candidates = candidates.OrderByDescending(featureImportanceFunc).ToList(); //order descending
return FilterScatteredFeatures(candidates, maxNumOfFeatures, 5); //candidates.Count must be >= MIN_NUM_OF_FEATURES
}
/// <summary>
/// Calculates phase using Atan2 (secondImage / firstImage).
/// </summary>
/// <param name="imageX">First image.</param>
/// <param name="imageY">Second image.</param>
/// <returns>Phase.</returns>
public static Gray<float>[,] Phase(this Gray<float>[,] imageX, Gray<float>[,] imageY)
{
return imageX.Calculate(imageY, phase_Float, inPlace: false);
}
/// <summary>
/// Adjusts pixels' contrast value by increasing RGB values of bright pixel and decreasing
/// pixel values of dark pixels (or vise versa if contrast needs to be decreased).
/// </summary>
/// <param name="im">Image.</param>
/// <param name="factor">Factor which is used to adjust contrast. Factor values greater than
/// 0 increase contrast making light areas lighter and dark areas darker. Factor values
/// less than 0 decrease contrast - decreasing variety of contrast.</param>
/// <param name="inPlace">Process in place or make not. If in place is set to true, returned value may be discarded.</param>
/// <returns>Corrected image.</returns>
public static Gray <byte>[,] CorrectContrast(this Gray <byte>[,] im, int factor = 10, bool inPlace = false)
{
ContrastCorrection conrastCorrection = new ContrastCorrection(factor);
return(im.ApplyFilter(conrastCorrection, inPlace));
}
/// <summary>
/// Process image looking for corners.
/// </summary>
/// <param name="cornerDetector">Corner detection algorithm instance.</param>
/// <param name="image">Source image to process.</param>
/// <returns>Returns list of found corners (X-Y coordinates).</returns>
public static List <Point> ProcessImage(this ICornersDetector cornerDetector, Gray <byte>[,] image)
{
List <Point> points = null;
using (var uImg = image.Lock())
{
points = cornerDetector.ProcessImage(uImg.AsAForgeImage())
.Select(x => x.ToPoint())
.ToList();
}
return(points);
}
/// <summary>
/// Camshift algorithm
/// </summary>
/// <param name="probabilityMap">Probability map [0-1].</param>
/// <param name="roi">Initial Search area</param>
/// <returns>Object position, size and angle packed into a structure.</returns>
public static Box2D Process(Gray<byte>[,] probabilityMap, Rectangle roi)
{
CentralMoments centralMoments;
return Process(probabilityMap, roi, Meanshift.DEFAULT_TERM, out centralMoments);
}
/// <summary>
/// Camshift algorithm
/// </summary>
/// <param name="probabilityMap">Probability map [0-255].</param>
/// <param name="roi">Initial Search area</param>
/// <param name="termCriteria">Mean shift termination criteria (PLEASE DO NOT REMOVE (but you can move it) THIS CLASS; PLEASE!!!)</param>
/// <param name="centralMoments">Calculated central moments (up to order 2).</param>
/// <returns>Object position, size and angle packed into a structure.</returns>
public static Box2D Process(Gray<byte>[,] probabilityMap, Rectangle roi, TermCriteria termCriteria, out CentralMoments centralMoments)
{
// Compute mean shift
Rectangle objArea = Meanshift.Process(probabilityMap, roi, termCriteria, out centralMoments);
//fit ellipse
Ellipse ellipse = centralMoments.GetEllipse();
ellipse.Center = objArea.Center();
//return empty structure is the object is lost
var sz = ellipse.Size;
if (Single.IsNaN(sz.Width) || Single.IsNaN(sz.Height) ||
sz.Width < 1 || sz.Height < 1)
{
return Box2D.Empty;
}
return (Box2D)ellipse;
}
/// <summary>
/// The blending filter is able to blend two images using a homography matrix.
/// A linear alpha gradient is used to smooth out differences between the two
/// images, effectively blending them in two images. The gradient is computed
/// considering the distance between the centers of the two images.
/// <para>Homography matrix is set to identity.</para>
/// <para>Fill color is set to black with alpha set to 0 (all zeros).</para>
/// </summary>
/// <param name="im">Image.</param>
/// <param name="overlayIm">The overlay image (also called the anchor).</param>
/// <param name="gradient">A value indicating whether to blend using a linear
/// gradient or just superimpose the two images with equal weights.</param>
/// <param name="alphaOnly">A value indicating whether only the alpha channel
/// should be blended. This can be used together with a transparency
/// mask to selectively blend only portions of the image.</param>
/// <returns>Blended image.</returns>
public static Bgra<byte>[,] Blend(this Gray<byte>[,] im, Gray<byte>[,] overlayIm, bool gradient = true, bool alphaOnly = false)
{
return Blend(im, overlayIm, new MatrixH(Matrix.Identity(3)), new Bgra<byte>(), gradient, alphaOnly);
}
/// <summary>
/// Calculates histogram.
/// </summary>
/// <param name="channel">Image channel.</param>
/// <param name="accumulate">Accumulate or erase histogram before.</param>
/// <param name="mask">Mask for image color locations.</param>
/// <param name="maskOffset">The location offset for the mask. The mask area will be [offsetX, offsetY, channelWidth, channelHeight].</param>
public void Calculate(Gray<byte>[,] channel, bool accumulate, Gray<byte>[,] mask, Point maskOffset)
{
Calculate(new Gray<byte>[][,]{ channel }, accumulate, mask, maskOffset);
}
/// <summary>
/// Creates template from the input image by using provided parameters.
/// </summary>
/// <param name="sourceImage">Input image.</param>
/// <param name="minFeatureStrength">Minimum gradient value for the feature.</param>
/// <param name="maxNumberOfFeatures">Maximum number of features per template. The features will be extracted so that their locations are semi-uniformly spread.</param>
/// <param name="classLabel">Template class label.</param>
public virtual void Initialize(Gray<byte>[,] sourceImage, int minFeatureStrength, int maxNumberOfFeatures, string classLabel)
{
Gray<int>[,] sqrMagImg;
Gray<int>[,] orientationImg = GradientComputation.Compute(sourceImage, out sqrMagImg, minFeatureStrength);
Func<Feature, int> featureImportanceFunc = (feature) => sqrMagImg[feature.Y, feature.X].Intensity;
Initialize(orientationImg, maxNumberOfFeatures, classLabel, featureImportanceFunc);
}
/// <summary>
/// Creates template from the input image by using provided parameters.
/// </summary>
/// <param name="orientation">Orientation image.</param>
/// <param name="maxNumberOfFeatures">Maximum number of features per template. The features will be extracted so that their locations are semi-uniformly spread.</param>
/// <param name="classLabel">Template class label.</param>
/// <param name="featureImportanceFunc">Function which returns feature's strength.</param>
public void Initialize(Gray<int>[,] orientation, int maxNumberOfFeatures, string classLabel, Func<Feature, int> featureImportanceFunc = null)
{
maxNumberOfFeatures = System.Math.Max(0, System.Math.Min(maxNumberOfFeatures, GlobalParameters.MAX_NUM_OF_FEATURES));
featureImportanceFunc = (featureImportanceFunc != null) ? featureImportanceFunc: (feature) => 0;
Gray<byte>[,] importantQuantizedOrient = FeatureMap.Calculate(orientation, 0);
List<Feature> features = ExtractTemplate(importantQuantizedOrient, maxNumberOfFeatures, featureImportanceFunc);
BoundingRect = GetBoundingRectangle(features);
//if (boundingRect.X == 1 && boundingRect.Y == 1 && boundingRect.Width == 18)
// Console.WriteLine();
for (int i = 0; i < features.Count; i++)
{
features[i].X -= BoundingRect.X;
features[i].Y -= BoundingRect.Y;
//if(features[i].X < 0 || features[i].Y < 0)
// Console.WriteLine();
//PATCH!!!
features[i].X = System.Math.Max(0, features[i].X);
features[i].Y = System.Math.Max(0, features[i].Y);
}
this.Features = features.ToArray();
this.Size = BoundingRect.Size;
this.ClassLabel = classLabel;
}
/// <summary>
/// Meanshift algorithm
/// </summary>
/// <param name="probabilityMap">Probability map [0-1].</param>
/// <param name="roi">Initial search area</param>
/// <returns>Object area.</returns>
public static Rectangle Process(Gray<byte>[,] probabilityMap, Rectangle roi)
{
CentralMoments centralMoments;
return process(probabilityMap, roi, DEFAULT_TERM, out centralMoments);
}
