public ImageService(IEventAggregator eventAggregator)
{
_eventAggregator = eventAggregator;
_eventAggregator.Subscribe(this);
_cascade = new FaceHaarCascade();
_harHaarObjectDetector = new HaarObjectDetector(_cascade, 30);
SetupDetector();
}
public void MaxSizeTest()
{
HaarCascade cascade = new FaceHaarCascade();
HaarObjectDetector target = new HaarObjectDetector(cascade,
50, ObjectDetectorSearchMode.Default);
Bitmap bmp = Properties.Resources.lena_color;
Rectangle[] result;
target.MaxSize = new Size(10, 60);
result = target.ProcessFrame(bmp);
Assert.AreEqual(0, result.Length);
target.MaxSize = new Size(60, 60);
result = target.ProcessFrame(bmp);
Assert.AreEqual(1, result.Length);
foreach (var r in result)
{
Assert.IsTrue(r.Width <= target.MaxSize.Width);
Assert.IsTrue(r.Height <= target.MaxSize.Height);
}
target.MaxSize = new Size(80, 80);
result = target.ProcessFrame(bmp);
Assert.AreEqual(2, result.Length);
foreach (var r in result)
{
Assert.IsTrue(r.Width <= target.MaxSize.Width);
Assert.IsTrue(r.Height <= target.MaxSize.Height);
}
}
public void ProcessFrame2()
{
HaarCascade cascade = new FaceHaarCascade();
HaarObjectDetector target = new HaarObjectDetector(cascade,
30, ObjectDetectorSearchMode.NoOverlap);
Bitmap bmp = Properties.Resources.lena_gray;
target.ProcessFrame(bmp);
Assert.AreEqual(1, target.DetectedObjects.Length);
Assert.AreEqual(255, target.DetectedObjects[0].X);
Assert.AreEqual(225, target.DetectedObjects[0].Y);
Assert.AreEqual(123, target.DetectedObjects[0].Width);
Assert.AreEqual(123, target.DetectedObjects[0].Height);
target = new HaarObjectDetector(cascade,
30, ObjectDetectorSearchMode.Default);
target.ProcessFrame(bmp);
Assert.AreEqual(6, target.DetectedObjects.Length);
Assert.AreEqual(255, target.DetectedObjects[0].X);
Assert.AreEqual(225, target.DetectedObjects[0].Y);
Assert.AreEqual(123, target.DetectedObjects[0].Width);
Assert.AreEqual(123, target.DetectedObjects[0].Height);
}
// Constructor
public MainForm()
{
InitializeComponent();
HaarCascade cascade = new FaceHaarCascade();
detector = new HaarObjectDetector(cascade,
25, ObjectDetectorSearchMode.Single, 1.2f,
ObjectDetectorScalingMode.GreaterToSmaller);
}
public void ParseTest()
{
StringReader stringReader = new StringReader(Properties.Resources.haarcascade_frontalface_alt);
HaarCascade cascade1 = HaarCascade.FromXml(stringReader);
Assert.AreEqual(22, cascade1.Stages.Length);
Assert.AreEqual(3, cascade1.Stages[0].Trees.Length);
Assert.AreEqual(1, cascade1.Stages[0].Trees[0].Length);
Assert.AreEqual(false, cascade1.HasTiltedFeatures);
// Load the hard coded version of the classifier
HaarCascade cascade2 = new FaceHaarCascade();
Assert.AreEqual(cascade1.Stages.Length, cascade2.Stages.Length);
Assert.AreEqual(cascade1.Height, cascade2.Height);
Assert.AreEqual(cascade1.Width, cascade2.Width);
for (int i = 0; i < 3; i++)
{
HaarCascadeStage stage1 = cascade1.Stages[i];
HaarCascadeStage stage2 = cascade2.Stages[i];
//Assert.AreEqual(stage1.NextIndex, stage2.NextIndex);
//Assert.AreEqual(stage1.ParentIndex, stage2.ParentIndex);
Assert.AreEqual(stage1.Threshold, stage2.Threshold);
Assert.AreEqual(stage1.Trees.Length, stage2.Trees.Length);
for (int j = 0; j < stage1.Trees.Length && j < stage2.Trees.Length; j++)
{
HaarFeatureNode[] tree1 = stage1.Trees[j];
HaarFeatureNode[] tree2 = stage2.Trees[j];
Assert.AreEqual(tree1.Length, tree2.Length);
for (int k = 0; k < tree1.Length; k++)
{
HaarFeatureNode node1 = tree1[k];
HaarFeatureNode node2 = tree2[k];
Assert.AreEqual(node1.LeftNodeIndex, node2.LeftNodeIndex);
Assert.AreEqual(node1.RightNodeIndex, node2.RightNodeIndex);
Assert.AreEqual(node1.LeftValue, node2.LeftValue);
Assert.AreEqual(node1.RightValue, node2.RightValue);
Assert.AreEqual(node1.Feature.Tilted, node2.Feature.Tilted);
Assert.AreEqual(node1.Threshold, node2.Threshold, 0.000000001);
}
}
}
}
public string CleanUpImage(string rawDataUrl, out string intermediateUrl)
{
intermediateUrl = null;
var cs = new FaceHaarCascade();
var detector = new HaarObjectDetector(cs, 30)
{
SearchMode = ObjectDetectorSearchMode.Average,
ScalingMode = ObjectDetectorScalingMode.SmallerToGreater,
ScalingFactor = 1.5f,
UseParallelProcessing = true,
Suppression = 2
};
string contentType;
var data = GetDataFromUrl(rawDataUrl, out contentType);
using (var ms = new MemoryStream(data))
{
var image = (Bitmap)Bitmap.FromStream(ms);
new ContrastStretch().ApplyInPlace(image);
var faces = detector.ProcessFrame(image);
if (faces.Length > 0)
{
var intermediateImage = new Bitmap(image);
new RectanglesMarker(faces, Color.Red).ApplyInPlace(intermediateImage);
var boundary = Math.Max(40, faces.Max(i => Math.Max(i.Height, i.Width)));
var x1 = Math.Max(0, faces.Min(i => i.Left) - boundary);
var y1 = Math.Max(0, faces.Min(i => i.Top) - boundary);
var x2 = Math.Min(image.Width, faces.Max(i => i.Right) + boundary);
var y2 = Math.Min(image.Height, faces.Max(i => i.Bottom) + boundary);
var newBoundingBox = new Rectangle(x1, y1, x2 - x1, y2 - y1);
new RectanglesMarker(new [] { newBoundingBox }, Color.Blue).ApplyInPlace(intermediateImage);
using (var ms2 = new MemoryStream())
{
intermediateImage.Save(ms2, ImageFormat.Jpeg);
intermediateUrl = string.Concat("data:image/jpeg;base64,", Convert.ToBase64String(ms2.ToArray()));
}
// perform no cropping of the image - post the original
}
// save off at JPG/100
var finalImage = ImageHelper.GetBytes(s => image.Save(s, ImageHelper.JPEGEncoder(), ImageHelper.Quality(100)));
var newDataUrl = string.Concat("data:image/jpeg;base64,", Convert.ToBase64String(finalImage));
return newDataUrl;
}
}
public MainForm()
{
InitializeComponent();
cbMode.DataSource = Enum.GetValues(typeof(ObjectDetectorSearchMode));
cbScaling.DataSource = Enum.GetValues(typeof(ObjectDetectorScalingMode));
cbMode.SelectedItem = ObjectDetectorSearchMode.Single;
cbScaling.SelectedItem = ObjectDetectorScalingMode.GreaterToSmaller;
toolStripStatusLabel1.Text = "Please select the detector options and click Detect to begin.";
HaarCascade cascade = new FaceHaarCascade();
detector = new HaarObjectDetector(cascade, 30);
}
public void ProcessFrame()
{
HaarCascade cascade = new FaceHaarCascade();
HaarObjectDetector target = new HaarObjectDetector(cascade,
50, ObjectDetectorSearchMode.NoOverlap);
Bitmap bmp = Properties.Resources.lena_color;
target.ProcessFrame(bmp);
Assert.AreEqual(1, target.DetectedObjects.Length);
Assert.AreEqual(126, target.DetectedObjects[0].X);
Assert.AreEqual(112, target.DetectedObjects[0].Y);
Assert.AreEqual(59, target.DetectedObjects[0].Width);
Assert.AreEqual(59, target.DetectedObjects[0].Height);
}
public void CloneTest()
{
HaarCascade expected = new FaceHaarCascade();
HaarCascade actual = (HaarCascade)expected.Clone();
Assert.AreNotEqual(expected, actual);
Assert.AreEqual(expected.HasTiltedFeatures, actual.HasTiltedFeatures);
Assert.AreEqual(expected.Height, actual.Height);
Assert.AreEqual(expected.Width, actual.Width);
Assert.AreNotEqual(expected.Stages, actual.Stages);
Assert.AreEqual(expected.Stages.Length, actual.Stages.Length);
for (int i = 0; i < expected.Stages.Length; i++)
equals(expected.Stages[i], actual.Stages[i]);
}
public void ProcessFrame()
{
#region doc_example
// In order to use a HaarObjectDetector, first we have to tell it
// which type of objects we would like to detect. And in a Haar detector,
// different object classifiers are specified in terms of a HaarCascade.
// The framework comes with some built-in cascades for common body
// parts, such as Face and Nose. However, it is also possible to
// load a cascade from cascade XML definitions in OpenCV 2.0 format.
// In this example, we will be creating a cascade for a Face detector:
var cascade = new Accord.Vision.Detection.Cascades.FaceHaarCascade();
// Note: In the case we would like to load it from XML, we could use:
// var cascade = HaarCascade.FromXml("filename.xml");
// Now, create a new Haar object detector with the cascade:
var detector = new HaarObjectDetector(cascade, minSize: 50,
searchMode: ObjectDetectorSearchMode.NoOverlap);
// Note that we have specified that we do not want overlapping objects,
// and that the minimum object an object can have is 50 pixels. Now, we
// can use the detector to classify a new image. For instance, consider
// the famous Lena picture:
Bitmap bmp = Accord.Imaging.Image.Clone(Resources.lena_color);
// We have to call ProcessFrame to detect all rectangles containing the
// object we are interested in (which in this case, is the face of Lena):
Rectangle[] rectangles = detector.ProcessFrame(bmp);
// The answer will be a single rectangle of dimensions
//
// {X = 126 Y = 112 Width = 59 Height = 59}
//
// which indeed contains the only face in the picture.
#endregion
Assert.AreEqual(1, detector.DetectedObjects.Length);
Assert.AreEqual(126, detector.DetectedObjects[0].X);
Assert.AreEqual(112, detector.DetectedObjects[0].Y);
Assert.AreEqual(59, detector.DetectedObjects[0].Width);
Assert.AreEqual(59, detector.DetectedObjects[0].Height);
}
static void Main(string[] args)
{
//http://www.codeproject.com/Tips/561129/Face-Detection-with-Lines-of-Code-VB-NET
//describing Viola Jones here : http://makematics.com/research/viola-jones/
//choosing scaling factor : http://www.mathworks.com/help/vision/ref/vision.cascadeobjectdetector-class.html#btc108o
string fileName = "9_r.jpg";
var image = new Bitmap("C:/temp/FaceDetection/"+fileName);
var cascade = new FaceHaarCascade();
var detector = new HaarObjectDetector(cascade,30);
detector.SearchMode = ObjectDetectorSearchMode.Average;
detector.Suppression = 3;
detector.MaxSize =new Size(image.Width,image.Height);
int scalingValue = image.Width > image.Height ? image.Width : image.Height;
detector.ScalingFactor = scalingValue / (scalingValue-0.5f);
detector.ScalingMode = ObjectDetectorScalingMode.GreaterToSmaller;
detector.UseParallelProcessing = true;
detector.Suppression = 1;
var sw = new Stopwatch();
sw.Start();
Rectangle[] faceObjects = detector.ProcessFrame(image);
var p = new Pen(Color.Aqua,10);
var graphicRect = Graphics.FromImage(image);
foreach (var face in faceObjects)
{
graphicRect.DrawRectangle(p, face);
}
graphicRect.Dispose();
image.Save("C:/temp/FaceDetection/Results/Average_3/"+fileName);
sw.Stop();
}
private async void MainWindow_Loaded(object sender, RoutedEventArgs e)
{
// Wait for the client to get setup
await Task.Delay(5000);
// Setup Socket Client
Client = new TcpClient();
Client.Connect(IPAddress.Parse("10.10.10.100"), 1911);
ClientStream = Client.GetStream();
// enumerate video devices
videoDevices = new FilterInfoCollection(FilterCategory.VideoInputDevice);
// create video source
videoSource = new VideoCaptureDevice(videoDevices[1].MonikerString);
videoSource.SnapshotResolution = videoSource.VideoCapabilities[7];
// set NewFrame event handler
videoSource.NewFrame += new NewFrameEventHandler(video_NewFrame);
// Setup Face Detection
Cascade = new FaceHaarCascade();
Detector = new HaarObjectDetector(Cascade, 30);
Detector.SearchMode = ObjectDetectorSearchMode.Average;
//Detector.ScalingFactor = 1f;
Detector.ScalingMode = ObjectDetectorScalingMode.GreaterToSmaller;
Detector.UseParallelProcessing = true;
Detector.Suppression = 3;
// Setup Tracking Data
CameraVector.X = 90;
CameraVector.Y = 90;
//ClientStream.Write(Encoding.UTF8.GetBytes("090,090"), 0, 7);
//await Task.Delay(3000);
// start the video source
videoSource.Start();
}
public void ProcessFrame3()
{
HaarCascade cascade = new FaceHaarCascade();
HaarObjectDetector target = new HaarObjectDetector(cascade,
15, ObjectDetectorSearchMode.NoOverlap);
Bitmap bmp = Properties.Resources.three;
target.ProcessFrame(bmp);
Assert.AreEqual(2, target.DetectedObjects.Length);
int i = 0;
/* Assert.AreEqual(180, target.DetectedObjects[0].X);
Assert.AreEqual(275, target.DetectedObjects[0].Y);
Assert.AreEqual(41, target.DetectedObjects[0].Width);
Assert.AreEqual(41, target.DetectedObjects[0].Height);
*/
Assert.AreEqual(168, target.DetectedObjects[i].X);
Assert.AreEqual(144, target.DetectedObjects[i].Y);
Assert.AreEqual(49, target.DetectedObjects[i].Width);
Assert.AreEqual(49, target.DetectedObjects[i].Height);
i++;
Assert.AreEqual(392, target.DetectedObjects[i].X);
Assert.AreEqual(133, target.DetectedObjects[i].Y);
Assert.AreEqual(59, target.DetectedObjects[i].Width);
Assert.AreEqual(59, target.DetectedObjects[i].Height);
target = new HaarObjectDetector(cascade,
15, ObjectDetectorSearchMode.Single);
target.ProcessFrame(bmp);
Assert.AreEqual(1, target.DetectedObjects.Length);
}
public void ParseTest3()
{
HaarCascade cascade = new FaceHaarCascade();
Assert.AreEqual(22, cascade.Stages.Length);
Assert.AreEqual(3, cascade.Stages[0].Trees.Length);
Assert.AreEqual(1, cascade.Stages[0].Trees[0].Length);
}
public void ProcessVideo()
{
string basePath = Path.Combine(NUnit.Framework.TestContext.CurrentContext.TestDirectory, "detector");
#region doc_video
// Let's test the detector using a sample video from
// the collection of test videos in the framework:
TestVideos ds = new TestVideos(basePath);
string fileName = ds["crowd.mp4"];
// In this example, we will be creating a cascade for a Face detector:
var cascade = new Accord.Vision.Detection.Cascades.FaceHaarCascade();
// Now, create a new Haar object detector with the cascade:
var detector = new HaarObjectDetector(cascade, minSize: 25,
searchMode: ObjectDetectorSearchMode.Average,
scalingMode: ObjectDetectorScalingMode.SmallerToGreater,
scaleFactor: 1.1f)
{
Suppression = 5 // This should make sure we only report regions as faces if
// they have been detected at least 5 times within different cascade scales.
};
// Now, let's open the video using FFMPEG:
var video = new VideoFileReader();
video.Open(fileName);
// And then check the contents of one of the frames:
Bitmap frame = video.ReadVideoFrame(frameIndex: 0);
// Creating bitmaps and locking them is an expensive
// operation. Instead, let's allocate once and reuse
BitmapData bitmapData = frame.LockBits(ImageLockMode.ReadWrite);
UnmanagedImage unmanagedImage = new UnmanagedImage(bitmapData);
// We will create a color marker to show the faces
var objectMarker = new RectanglesMarker(Color.Red);
// This example is going to show two different ways to save results to disk. The
// first is to save the results frame-by-frame, saving each individual frame as
// a separate .png file. The second is to save them as a video in .mp4 format.
// To save results as a movie clip in mp4 format, you can use:
VideoFileWriter writer = new VideoFileWriter();
writer.Open(Path.Combine(basePath, "detected_faces.mp4"), frame.Width, frame.Height);
// Now, for each frame of the video
for (int frameIndex = 0; frameIndex < video.FrameCount; frameIndex++)
{
// Read the current frame into the bitmap data
video.ReadVideoFrame(frameIndex, bitmapData);
// Feed the frame to the tracker
Rectangle[] faces = detector.ProcessFrame(unmanagedImage);
// Mark the location of the tracker object in red color
objectMarker.Rectangles = faces;
objectMarker.ApplyInPlace(unmanagedImage); // overwrite the frame
// Save it to disk: first saving each frame separately:
frame.Save(Path.Combine(basePath, "frame_{0}.png".Format(frameIndex)));
// And then, saving as a .mp4 file:
writer.WriteVideoFrame(bitmapData);
}
// The generated video can be seen at https://1drv.ms/v/s!AoiTwBxoR4OAoLJhPozzixD25XcbiQ
video.Close();
writer.Close();
#endregion
}
