public static IImage ToImage(this CvImage input, PixelFormat pixelFormat, bool takeOwnership)
{
if (input == CvImage.Damon)
{
takeOwnership = false;
}
return(new MatImage(input.Mat, pixelFormat, takeOwnership));
}
public static CvImage EnforceGrayDefault(CvImage img) => img != Damon ? img : Gray;
private static string model = "yolov2-voc.weights"; //YOLOv2 544x544
public static YOLODescriptor Detect(CvImage image, float threshold, bool enabled)
{
SpreadBuilder <Rect> rectSB = Spread.CreateBuilder <Rect>();
SpreadBuilder <float> confidenceSB = Spread.CreateBuilder <float>();
SpreadBuilder <int> detectedClassSB = Spread.CreateBuilder <int>();
SpreadBuilder <float> classProbabilitySB = Spread.CreateBuilder <float>();
SpreadBuilder <string> labelSB = Spread.CreateBuilder <string>();
if (enabled && image != CvImage.Damon)
{
var w = image.Width;
var h = image.Height;
//setting blob, parameter are important
var blob = CvDnn.BlobFromImage(image.Mat, 1 / 255.0, new Size(544, 544), new Scalar(), true, false);
if (net == null || net.IsDisposed)
{
net = CvDnn.ReadNetFromDarknet(cfg, model);
}
net.SetInput(blob, "data");
//forward model
var prob = net.Forward();
/* YOLO2 VOC output
* 0 1 : center 2 3 : w/h
* 4 : confidence 5 ~24 : class probability */
const int prefix = 5; //skip 0~4
for (int i = 0; i < prob.Rows; i++)
{
var matchConfidence = prob.At <float>(i, 4);
if (matchConfidence > threshold)
{
//get classes probability
Point min, max;
Cv2.MinMaxLoc(prob.Row[i].ColRange(prefix, prob.Cols), out min, out max);
var classes = max.X;
var probability = prob.At <float>(i, classes + prefix);
if (probability > threshold) //more accuracy
{
//get center and width/height
var centerX = prob.At <float>(i, 0) * w;
var centerY = prob.At <float>(i, 1) * h;
var width = prob.At <float>(i, 2) * w;
var height = prob.At <float>(i, 3) * h;
var x1 = (centerX - width / 2) < 0 ? 0 : centerX - width / 2; //avoid left side over edge
//org.Rectangle(new Point(centerX - width / 2, centerY - height / 2), new Point(centerX + width / 2, centerY + height / 2), Colors[classes], 2);
rectSB.Add(new Rect(new Point(x1, centerY - height / 2), new Size(width, height)));
confidenceSB.Add(matchConfidence);
detectedClassSB.Add(classes);
classProbabilitySB.Add(probability);
labelSB.Add(Labels[classes]);
}
}
}
}
YOLODescriptor result = new YOLODescriptor();
result.confidence = confidenceSB.ToSpread();
result.detectedClass = detectedClassSB.ToSpread();
result.classProbability = classProbabilitySB.ToSpread();
result.classLabel = labelSB.ToSpread();
result.rectangles = rectSB.ToSpread <Rect>();
return(result);
}
