//彩色图片直方图均衡化
public static Mat EqualizeHistForColorImage(Mat src)
{
Mat Image = new Mat(src.Size(), src.Depth(), 3);
int max_channel = 4;
Mat[] ImageChannel = new Mat[4];
int i = 0;
for (i = 0; i < src.Channels(); i++)
{
ImageChannel[i] = new Mat(src.Size(), src.Depth(), 1);
}
Cv2.Split(src, out ImageChannel);
for (i = 0; i < src.Channels(); i++)
{
Cv2.EqualizeHist(ImageChannel[i], ImageChannel[i]);
}
Cv2.Merge(ImageChannel, Image);
return(Image);
}
public void Detect(Mat frame)
{
var grayImage = new Mat();
Cv2.CvtColor(frame, grayImage, ColorConversionCodes.BGRA2GRAY);
Cv2.EqualizeHist(grayImage, grayImage);
var bodies = cascade.DetectMultiScale(
image: frame,
scaleFactor: 1.1,
minNeighbors: 2,
flags: HaarDetectionType.DoRoughSearch | HaarDetectionType.ScaleImage,
minSize: new Size(30, 30)
);
//return bodies;
Console.WriteLine("Detected bodies: {0}", bodies.Length);
foreach (var faceRect in bodies)
{
Cv2.Rectangle(frame, faceRect, Scalar.Red, 2);
}
}
//Face detection
private void DetectAndSaveImg(Mat image)
{
//Image to gray scale
using (var grayImage = new Mat()) {
Cv2.CvtColor(image, grayImage, ColorConversionCodes.BGR2GRAY);
//Flattening the histogram
Cv2.EqualizeHist(grayImage, grayImage);
//Face recognition, Small faces excluded
var mats = Cascade.DetectMultiScale(grayImage, 1.1, 3, 0, new Size(80, 80))
//Make rects focusing on facial parts
.Select(rect => new Rect(rect.X, rect.Y, rect.Width, rect.Height))
//Imaged cut out
.Select(image.Clone)
//Listing
.ToList();
SaveImg(mats);
}
}
public override void Apply(Mat input)
{
_start = DateTime.Now;
Input = input;
if (IsActive)
{
Mat equ = new Mat();
Cv2.CvtColor(input, equ, ColorConversionCodes.BGR2YCrCb);
Mat[] Channels;
Cv2.Split(equ, out Channels);
Cv2.EqualizeHist(Channels[0], Channels[0]);
Cv2.Merge(Channels, equ);
Cv2.CvtColor(equ, Output, ColorConversionCodes.YCrCb2BGR);
}
else
{
Cv2.CopyTo(Input, Output);
}
base.Apply(input);
}
private void button9_Click(object sender, EventArgs e)
{
Mat OriginalImage = new Mat(fileName, ImreadModes.Color);
Mat hsvIm = new Mat();
Cv2.CvtColor(OriginalImage, hsvIm, ColorConversionCodes.BGR2HSV);
Mat[] hsv;
hsv = Cv2.Split(hsvIm);
Cv2.EqualizeHist(hsv[2], hsv[2]);
Mat res = new Mat();
Cv2.Merge(hsv, res);
Bitmap bp = OpenCvSharp.Extensions.BitmapConverter.ToBitmap(hsv[2]);
createHistV(bp);
Cv2.CvtColor(res, res, ColorConversionCodes.HSV2BGR);
showFilter(res, "brightness equalize");
}
public static List <OpenCvSharp.Rect> DetectFaces(OpenCvSharp.Mat image)
{
List <OpenCvSharp.Rect> faces = new List <OpenCvSharp.Rect>();
var facesCascade = HttpContext.Current.Server.MapPath("~/face.xml");
using (OpenCvSharp.CascadeClassifier face = new OpenCvSharp.CascadeClassifier(facesCascade))
{
using (OpenCvSharp.Mat ugray = new OpenCvSharp.Mat())
{
Cv2.CvtColor(image, ugray, ColorConversionCodes.BGRA2GRAY);
Cv2.EqualizeHist(ugray, ugray);
var facesDetected = face.DetectMultiScale(
image: ugray,
scaleFactor: 1.1,
minNeighbors: 10,
flags: HaarDetectionType.DoRoughSearch | HaarDetectionType.ScaleImage,
minSize: new OpenCvSharp.Size(20, 20));
faces.AddRange(facesDetected);
}
}
return(faces);
}
private Rect[] FaceDetect(Mat source_image)
{
// 그레이 스케일
Mat frame_img_gray = new Mat();
byte[] imageBytes = source_image.ToBytes(".bmp");
//byte[] imageBytes = frame_img_FlipY.ToBytes(".bmp");
frame_img_gray = Mat.FromImageData(imageBytes, ImreadModes.Grayscale);
Cv2.EqualizeHist(frame_img_gray, frame_img_gray);
var cascade = new CascadeClassifier("C://Users//dev-yym//source//repos//opencv_003//FaceML_Data//haarcascade_frontalface_alt.xml");
//var nestedCascade = new CascadeClassifier("C://Users//dev-yym//source//repos//opencv_002//FaceML_Data//haarcascade_eye_tree_eyeglasses.xml");
var faces = cascade.DetectMultiScale(
image: frame_img_gray,
scaleFactor: 1.1,
minNeighbors: 2,
flags: HaarDetectionType.DoRoughSearch | HaarDetectionType.ScaleImage,
minSize: new OpenCvSharp.Size(30, 30)
);
return(faces);
}
private void button1_Click(object sender, EventArgs e)
{
video = new VideoCapture("D:\\project\\c#\\Day021\\vtest.avi");
Random r = new Random();
int cnt = 0;
int x = 0, y = 0, W = 0, H = 0;
int sleepTime = (int)Math.Round(1000 / video.Fps);
String filenameBodyCascade = "D:\\project\\c#\\Day021\\haarcascade_fullbody.xml"; // haar-cascade 훈련 사용
CascadeClassifier bodyCascade = new CascadeClassifier();
if (!bodyCascade.Load(filenameBodyCascade))
{
Console.WriteLine("error");
return;
}
inCvImage = new Mat();
outCvImage = new Mat();
int oH = inCvImage.Height;
int oW = inCvImage.Width;
bool csi = false;
while (true)
{
cnt = 0;
video.Read(inCvImage);
if (inCvImage.Empty())
{
break;
}
// detect
Rect[] body = bodyCascade.DetectMultiScale(inCvImage);
Console.WriteLine(body.Length);
foreach (var item in body)
{
Scalar c = new Scalar(r.Next(0, 255), r.Next(0, 255), r.Next(0, 255));
Cv2.Rectangle(inCvImage, item, c); // add rectangle to the image
Console.WriteLine("body : " + item);
cnt++;
x = item.X;
y = item.Y;
W = item.Width;
H = item.Height;
}
if (cnt > 4)
{
csi = true;
}
if (csi)
{
outCvImage = Mat.Ones(new OpenCvSharp.Size(oW, oH), MatType.CV_8UC1);
Cv2.CvtColor(inCvImage, outCvImage, ColorConversionCodes.BGR2GRAY);
Cv2.AdaptiveThreshold(outCvImage, outCvImage, 255,
AdaptiveThresholdTypes.GaussianC, ThresholdTypes.Binary, 25, 5);
}
else
{
outCvImage = inCvImage;
}
// display
//Cv2.ImShow("CCTV", frame);
picboxCow.ImageIpl = outCvImage;
if (Cv2.WaitKey(9) == 27)
{
inCvImage.Dispose(); outCvImage.Dispose();
video.Release();
break;
}
if (cnt > 7)
{
video.Release();
break;
}
}
// grayscale
Cv2.CvtColor(inCvImage, outCvImage, ColorConversionCodes.BGR2GRAY);
picboxCow.ImageIpl = outCvImage;
Delay(2000); //각 변환 시각화
//histogram equalizition을 통해서 픽셀를 고르게 분포
Cv2.EqualizeHist(outCvImage, outCvImage);
picboxCow.ImageIpl = outCvImage;
Delay(2000); //각 변환 시각화
dst1 = new Mat(); //누적 image 작업을 위한 초기화
dst2 = new Mat();
Cv2.Resize(outCvImage, dst1, new OpenCvSharp.Size(1920, 1280), 0, 0, InterpolationFlags.Lanczos4);
//무리 내의 사람 판독을 위해 따로 배출
Rect rect = new Rect(x, y, W, H);
dst1 = outCvImage.SubMat(rect);
//image 크기와 맞게 picplcsi 설정
picplCSi.Size = new System.Drawing.Size(W, H);
picplCSi.ImageIpl = dst1;
lbf.Visible = true;
Delay(2000);//각 변환 시각화
//보다 효율적으로 시각화 하기위한 사이즈 확대
Cv2.Resize(dst1, dst2, new OpenCvSharp.Size(800, 600), 0, 0, InterpolationFlags.Lanczos4);
//image 크기와 맞게 picplcsi 설정
this.Size = new System.Drawing.Size();
picplCSi.Size = new System.Drawing.Size(800, 600);
Cv2.ImShow("dst2", dst2);//image 출력
Cv2.WaitKey(2000);
//파일 내로 저장 후 메모리 누수 발생 방지를 위해 사용했던 Matrix Dispose시킵니다.
string _saveName = "C:/images/" + DateTime.Now.ToString("yyyy/MM/dd_hh_mm_ss") + ".jpeg";
Cv2.ImWrite(_saveName, dst2);
inCvImage.Dispose();
outCvImage.Dispose();
dst1.Dispose();
//Check Message
MessageBox.Show("업로드", "Checking", MessageBoxButtons.OK, MessageBoxIcon.Information);
Close();
}
static void Main(string[] args)
{
var detectedFaceGrayImage = new Mat();
var grayImage = new Mat();
var srcImage = new Mat();
var cascade = new CascadeClassifier(@"..\..\Data\haarcascade_frontalface_alt.xml");
var nestedCascade = new CascadeClassifier(@"..\..\Data\haarcascade_eye_tree_eyeglasses.xml");
var smile = new CascadeClassifier(@"..\..\Data\haarcascade_smile.xml");
VideoCapture video = new VideoCapture(0);
for (int i = 0; i < 1000; i++)
{
video.Read(srcImage);
Cv2.CvtColor(srcImage, grayImage, ColorConversionCodes.BGRA2GRAY);
Cv2.EqualizeHist(grayImage, grayImage);
var faces = cascade.DetectMultiScale(
image: grayImage,
scaleFactor: 1.1,
minNeighbors: 2,
flags: HaarDetectionType.DoRoughSearch | HaarDetectionType.ScaleImage,
minSize: new Size(30, 30)
);
foreach (var faceRect in faces)
{
var detectedFaceImage = new Mat(srcImage, faceRect);
Cv2.Rectangle(srcImage, faceRect, Scalar.Red, 2);
var nestedObjects = nestedCascade.DetectMultiScale(
image: grayImage,
scaleFactor: 1.1,
minNeighbors: 2,
flags: HaarDetectionType.DoRoughSearch | HaarDetectionType.ScaleImage,
minSize: new Size(30, 30)
);
foreach (var nestedObject in nestedObjects)
{
Cv2.Rectangle(srcImage, nestedObject, Scalar.YellowGreen, 2);
}
var nestedSmile = smile.DetectMultiScale(
image: grayImage,
scaleFactor: 3.1,
minNeighbors: 2,
flags: HaarDetectionType.DoRoughSearch | HaarDetectionType.ScaleImage,
minSize: new Size(30, 30)
);
foreach (var nestedObject in nestedSmile)
{
Cv2.Rectangle(srcImage, nestedObject, Scalar.Green, 2);
}
}
Cv2.ImShow("tela", srcImage);
Cv2.WaitKey(1); // do events
}
Cv2.WaitKey(0);
Cv2.DestroyAllWindows();
srcImage.Dispose();
}
// Simple example of canny edge detect
void ProcessImage(Mat _image)
{
Cv2.Flip(_image, _image, FlipMode.X);
Cv2.Canny(_image, cannyImage, 100, 100);
var grayImage = new Mat();
Cv2.CvtColor(_image, grayImage, ColorConversionCodes.BGR2GRAY);
Cv2.EqualizeHist(grayImage, grayImage);
var face_cascade = new CascadeClassifier();
face_cascade.Load(Application.dataPath + "/Plugins/Classifiers/haarcascade_frontalface_default.xml");
var eye_cascade = new CascadeClassifier();
eye_cascade.Load(Application.dataPath + "/Plugins/Classifiers/haarcascade_eye_tree_eyeglasses.xml");
//var right_ear_cascade = new CascadeClassifier();
//right_ear_cascade.Load(Application.dataPath + "/Plugins/Classifiers/haarcascade_mcs_leftear.xml");
//var left_ear_cascade = new CascadeClassifier();
//left_ear_cascade.Load(Application.dataPath + "/Plugins/Classifiers/haarcascade_mcs_rightear.xml");
//var mouth_cascade = new CascadeClassifier();
//mouth_cascade.Load(Application.dataPath + "/Plugins/Classifiers/haarcascade_mcs_mouth.xml");
//Debug.Log(" ");
var faces = face_cascade.DetectMultiScale(
image: grayImage,
scaleFactor: 1.3,
minNeighbors: 5,
flags: HaarDetectionType.DoRoughSearch | HaarDetectionType.ScaleImage,
minSize: new Size(100, 100)
);
Bounds meshRendererBounds = GetComponentInChildren <MeshRenderer>().bounds;
Vector3 meshRendererCenter = meshRendererBounds.center;
Vector3 maxBound = meshRendererBounds.max;
Vector3 minBound = meshRendererBounds.min;
OpenCvSharp.Rect rect = new OpenCvSharp.Rect((int)meshRendererCenter.x + 350, (int)meshRendererCenter.y + 50, 600, 600);
var global_rectangle_color = Scalar.FromRgb(0, 0, 255);
Cv2.Rectangle(_image, rect, global_rectangle_color, 3);
//Console.WriteLine("Detected faces: {0}", faces.Length);
//Debug.Log(faces.Length);
//var rnd = new System.Random();
var face_count = 0;
var eye_count = 0;
//var ear_count = 0;
//var mouth_count = 0;
foreach (var faceRect in faces)
{
var detectedFaceImage = new Mat(_image, faceRect);
//Cv2.ImShow(string.Format("Face {0}", face_count), detectedFaceImage);
//Cv2.WaitKey(1); // do events
var facec_rectangle_color = Scalar.FromRgb(255, 0, 0);
Cv2.Rectangle(_image, faceRect, facec_rectangle_color, 3);
var detectedFaceGrayImage = new Mat();
Cv2.CvtColor(detectedFaceImage, detectedFaceGrayImage, ColorConversionCodes.BGRA2GRAY);
var eyes = eye_cascade.DetectMultiScale(
image: grayImage,
scaleFactor: 1.3,
minNeighbors: 5,
flags: HaarDetectionType.DoRoughSearch | HaarDetectionType.ScaleImage,
minSize: new Size(50, 50)
);
foreach (var eyeRect in eyes)
{
var detectedEyeImage = new Mat(_image, eyeRect);
//Cv2.ImShow(string.Format("Face {0}", eye_count), detectedEyeImage);
//Cv2.WaitKey(1); // do events
var eye_rectangle_color = Scalar.FromRgb(0, 255, 0);
Cv2.Rectangle(_image, eyeRect, eye_rectangle_color, 3);
var detectedEyeGrayImage = new Mat();
Cv2.CvtColor(detectedEyeImage, detectedEyeGrayImage, ColorConversionCodes.BGRA2GRAY);
eye_count++;
}
//var left_ear = left_ear_cascade.DetectMultiScale(
// image: grayImage,
// scaleFactor: 1.3,
// minNeighbors: 5,
// flags: HaarDetectionType.DoRoughSearch | HaarDetectionType.ScaleImage,
// minSize: new Size(50, 50)
//);
//foreach (var ear in left_ear)
//{
// var detectedEarImage = new Mat(_image, ear);
// //Cv2.ImShow(string.Format("Face {0}", eye_count), detectedEyeImage);
// //Cv2.WaitKey(1); // do events
// var eye_rectangle_color = Scalar.FromRgb(0, 255, 0);
// Cv2.Rectangle(_image, ear, eye_rectangle_color, 3);
// var detectedEyeGrayImage = new Mat();
// Cv2.CvtColor(detectedEarImage, detectedEyeGrayImage, ColorConversionCodes.BGRA2GRAY);
// ear_count++;
//}
// var right_ear = right_ear_cascade.DetectMultiScale(
// image: grayImage,
// scaleFactor: 1.3,
// minNeighbors: 5,
// flags: HaarDetectionType.DoRoughSearch | HaarDetectionType.ScaleImage,
// minSize: new Size(50, 50)
//);
//foreach (var ear in right_ear)
//{
// var detectedEarImage = new Mat(_image, ear);
// //Cv2.ImShow(string.Format("Face {0}", eye_count), detectedEyeImage);
// //Cv2.WaitKey(1); // do events
// var eye_rectangle_color = Scalar.FromRgb(0, 255, 0);
// Cv2.Rectangle(_image, ear, eye_rectangle_color, 3);
// var detectedEyeGrayImage = new Mat();
// Cv2.CvtColor(detectedEarImage, detectedEyeGrayImage, ColorConversionCodes.BGRA2GRAY);
// ear_count++;
//}
// var mouth = mouth_cascade.DetectMultiScale(
// image: grayImage,
// scaleFactor: 1.3,
// minNeighbors: 5,
// flags: HaarDetectionType.DoRoughSearch | HaarDetectionType.ScaleImage,
// minSize: new Size(50, 50)
//);
//foreach (var m in mouth)
//{
// var detectedEarImage = new Mat(_image, m);
// //Cv2.ImShow(string.Format("Face {0}", eye_count), detectedEyeImage);
// //Cv2.WaitKey(1); // do events
// var eye_rectangle_color = Scalar.FromRgb(0, 255, 0);
// Cv2.Rectangle(_image, m, eye_rectangle_color, 3);
// var detectedEyeGrayImage = new Mat();
// Cv2.CvtColor(detectedEarImage, detectedEyeGrayImage, ColorConversionCodes.BGRA2GRAY);
// mouth_count++;
//}
face_count++;
}
//Debug.Log(face_count);
if (face_count == 1 && eye_count == 2 && !waitSoundEffect)
{
//Debug.Log(faces[0]);
//Debug.Log(meshRendererCenter.x);
//Debug.Log((int)meshRendererCenter.y + 50);
Point origin = faces[0].Location;
float width = faces[0].Width;
float height = faces[0].Height;
// Verification si le rect de la face est bien dans la zone de photo
if (origin.X > (int)meshRendererCenter.x + 350 &&
origin.X + width < (int)meshRendererCenter.x + 350 + 600 &&
origin.Y > (int)meshRendererCenter.y + 50 &&
origin.Y + height < (int)meshRendererCenter.y + 5 + 600 &&
width > 400 &&
height > 400)
{
Debug.Log("Take photo !");
TakePhoto();
soundEffects.MakePhotoSound();
waitSoundEffect = true;
}
}
//Cv2.ImShow("Haar Detection", _image);
//Cv2.WaitKey(1); // do events
}
public static void EqualizeHistogram(this Mat self, Mat output)
{
Cv2.EqualizeHist(self, output);
}
static void Main(string[] args)
{
Console.WriteLine("Trying to capture video");
// Opens MP4 file (ffmpeg is probably needed)
using var capture = new VideoCapture(0);
int sleepTime = (int)Math.Round(1000 / capture.Fps);
using (var window = new Window("capture"))
{
// Frame image buffer
var image = new Mat();
// When the movie playback reaches end, Mat.data becomes NULL.
while (true)
{
capture.Read(image); // same as cvQueryFrame
if (image.Empty())
{
break;
}
var grayImage = new Mat();
Cv2.CvtColor(image, grayImage, ColorConversionCodes.BGRA2GRAY);
Cv2.EqualizeHist(grayImage, grayImage);
var cascadeFrontalFace = new CascadeClassifier(@$ ".{Path.DirectorySeparatorChar}CascadeClassifiers{Path.DirectorySeparatorChar}haarcascade_frontalface_alt2.xml");
var cascade = new CascadeClassifier(@$ ".{Path.DirectorySeparatorChar}CascadeClassifiers{Path.DirectorySeparatorChar}haarcascade_profileface.xml");
var facesProfile = cascade.DetectMultiScale(
image: image,
scaleFactor: 1.1,
minNeighbors: 2,
flags: HaarDetectionType.DoRoughSearch | HaarDetectionType.ScaleImage,
minSize: new Size(30, 30)
);
var faces = cascadeFrontalFace.DetectMultiScale(
image: grayImage,
scaleFactor: 1.1,
minNeighbors: 2,
flags: HaarDetectionType.DoRoughSearch | HaarDetectionType.ScaleImage,
minSize: new Size(30, 30)
);
Console.WriteLine("{0} Detected faces: {1}", DateTime.Now, faces.Length + facesProfile.Length);
faces.ToList().ForEach(e => e.Height += 10);
facesProfile.ToList().ForEach(e => e.Height += 10);
foreach (var faceRect in faces)
{
Cv2.Rectangle(image, faceRect, Scalar.Red, 2);
}
foreach (var faceRect in facesProfile)
{
Cv2.Rectangle(image, faceRect, Scalar.Blue, 2);
}
window.ShowImage(image);
Cv2.WaitKey(sleepTime);
}
}
}
/*
* SK: Metóda, ktorá sa zavolá, pre vygenerovanie otázky
* ENG: Method, which is called for generating question
*/
void GenerateQuestion()
{
if (QnA.Count > 0)
{
currentQuestion = Random.Range(0, QnA.Count);
QuestionText.text = QnA[currentQuestion].Questiion;
Image.texture = QnA[currentQuestion].sample;
Mat image = Unity.TextureToMat(QnA[currentQuestion].sample);
// Deteguje tváre
var gray = image.CvtColor(ColorConversionCodes.BGR2GRAY);
Cv2.EqualizeHist(gray, gray);
// deteguje zhodné regióny (Faces bounding)
FileStorage storageFaces = new FileStorage(Faces.text, FileStorage.Mode.Read | FileStorage.Mode.Memory);
cascadeFaces = new CascadeClassifier();
if (!cascadeFaces.Read(storageFaces.GetFirstTopLevelNode()))
{
throw new System.Exception("FaceProcessor.Initialize: Failed to load Faces cascade classifier");
}
recognizer = FaceRecognizer.CreateFisherFaceRecognizer();
recognizer.Load(new FileStorage(RecognizerXml.text, FileStorage.Mode.Read | FileStorage.Mode.Memory));
// popisky (labels)
names = new string[] { "Cooper", "DeGeneres", "Nyongo", "Pitt", "Roberts", "Spacey" };
OpenCvSharp.Rect[] rawFaces = cascadeFaces.DetectMultiScale(gray, 1.1, 6);
foreach (var faceRect in rawFaces)
{
var grayFace = new Mat(gray, faceRect);
if (requiredSize.Width > 0 && requiredSize.Height > 0)
{
grayFace = grayFace.Resize(requiredSize);
}
int label = -1;
/*SK:
* pokús sa rozpoznať tvár:
* confidence" je v princípe vzdialenosť od vzorky k najbližšej známej tvári
* 0 je nejaká „ideálna zhoda“
*/
/*ENG:
* now try to recognize the face:
* confidence" here is actually a misguide. in fact, it's "distance from the sample to the closest known face".
* 0 being some "ideal match"
*/
double confidence = 0.0;
recognizer.Predict(grayFace, out label, out confidence);
faceName = names[label];
int line = 0;
const int textPadding = 2;
const double textScale = 2.0;
string messge = String.Format("{0}", names[label], (int)confidence);
var textSize = Cv2.GetTextSize(messge, HersheyFonts.HersheyPlain, textScale, 1, out line);
var textBox = new OpenCvSharp.Rect(
faceRect.X + (faceRect.Width - textSize.Width) / 2 - textPadding,
faceRect.Bottom,
textSize.Width + textPadding * 2,
textSize.Height + textPadding * 2
);
faceName = names[label];
Debug.Log(faceName);
}
// Priradenie obrázku textúre Image komponentu na scéne
// Asign image to the texture on Image component
var texture = Unity.MatToTexture(image);
var rawImage = Image;
rawImage.texture = texture;
var transform = Image.GetComponent <RectTransform>();
transform.sizeDelta = new Vector2(image.Width, image.Height);
for (int i = 0; i < Options.Length; i++)
{
Options[i].transform.GetChild(0).GetComponent <Text>().text = QnA[currentQuestion].Answers[i];
if (faceName == Options[i].transform.GetChild(0).GetComponent <Text>().text)
{
Options[i].GetComponent <AnswerScript>().isCorrect = true;
}
else
{
Options[i].GetComponent <AnswerScript>().isCorrect = false;
}
}
}
else
{
GameOver();
}
}
//展示车牌
private void ProcessAndShowImage(Bitmap image, PlateLocator.ParameterList pl)
{
currentTabCount = 0;
Mat matIn = image.ToMat();
AddTag("原图", matIn);
//转为hsv图片
Mat matHsv = matIn.CvtColor(ColorConversionCodes.BGR2HSV);
AddTag("HSV", matHsv);
//对v均衡化后在合并
Mat[] matToHsv = new Mat[3];
Cv2.Split(matHsv, out matToHsv);
Cv2.EqualizeHist(matToHsv[2], matToHsv[2]);
Mat equalizeHistHsv = new Mat();
Cv2.Merge(matToHsv, equalizeHistHsv);
AddTag("均衡HSV", equalizeHistHsv);
//在均衡化后的hsv颜色空间红寻找黄色和蓝色区域
Mat matYellow = new Mat();
Mat matBlue = new Mat();
Scalar yellow_low = new Scalar(15, 70, 70);
Scalar yellow_up = new Scalar(40, 255, 255);
Scalar blue_low = new Scalar(100, 70, 70);
Scalar blue_up = new Scalar(140, 255, 255);
Cv2.InRange(equalizeHistHsv, yellow_low, yellow_up, matYellow);
Cv2.InRange(equalizeHistHsv, blue_low, blue_up, matBlue);
Mat matAll = matBlue + matYellow;
AddTag("黄区", matYellow);
AddTag("蓝区", matBlue);
AddTag("黄蓝区", matAll);
//使用形态学操作对选定颜色区域进行处理
Mat matAllDilate = new Mat();
Mat matAllErode = new Mat();
Mat element = Cv2.GetStructuringElement(MorphShapes.Rect, new OpenCvSharp.Size(7, 3));
Cv2.Dilate(matAll, matAllDilate, element);
Cv2.Erode(matAllDilate, matAllErode, element);
AddTag("闭操作", matAllErode);
//寻找轮廓
OpenCvSharp.Point[][] contours; //vector<vector<Point>> contours;
HierarchyIndex[] hierarchyIndexes; //vector<Vec4i> hierarchy;
Cv2.FindContours(
matAllErode,
out contours,
out hierarchyIndexes,
mode: RetrievalModes.Tree,
method: ContourApproximationModes.ApproxSimple); //求轮廓
Mat matContours = matIn.Clone();
Cv2.DrawContours(matContours, contours, -1, new Scalar(0, 0, 255), 2); //画轮廓线
AddTag("轮廓", matContours);
//求轮廓的最小外接矩形
Mat matRects = matIn.Clone();
List <Rect> rects = new List <Rect>();
foreach (OpenCvSharp.Point[] p in contours)
{
Rect rect = Cv2.BoundingRect(p);
if ((double)rect.Height / rect.Width > pl.HeightDivideWidthLow && (double)rect.Height / rect.Width < pl.HeightDivedeWidthUp &&
rect.Height > pl.HeightLow && rect.Height < pl.HeightUp &&
rect.Width > pl.WidthLow && rect.Width < pl.WidthUp)
{
rects.Add(rect);
Cv2.Rectangle(matRects, rect, new Scalar(0, 0, 255), 3);
}
}
AddTag("外接矩形", matRects);
ShowSpliteImage(rects, matIn);
}
// 主要内容，图像处理方法的api
private Mat myOPENCV_run(Mat image_in, Mat image_out)
{
image_out = image_in; // 入图传给出图
for (int i = 0; i < listBox2.Items.Count; i++) //执行 列表框2内的方法
{
switch ((MyOPENCV)myOPENCV_runlist[i, 0]) // 列表框2内的运行方法
{
case MyOPENCV.cvt_color: //颜色转换 （入图，出图，颜色转换符，）
{
Cv2.CvtColor(image_out, image_out, (ColorConversionCodes)myOPENCV_runlist[i, 1], myOPENCV_runlist[i, 2]);
break;
}
case MyOPENCV.boxfilter: //方框滤波
{
OpenCvSharp.Size size;
size.Width = myOPENCV_runlist[i, 2];
size.Height = myOPENCV_runlist[i, 3];
Cv2.BoxFilter(image_out, image_out, myOPENCV_runlist[i, 1], size);
break;
}
case MyOPENCV.blur: //均值滤波
{
OpenCvSharp.Size size;
size.Width = myOPENCV_runlist[i, 1];
size.Height = myOPENCV_runlist[i, 2];
Cv2.Blur(image_out, image_out, size);
break;
}
case MyOPENCV.gaussianblur: // 高斯滤波
{
OpenCvSharp.Size size;
double sigmaX, sigmaY;
size.Width = myOPENCV_runlist[i, 1];
size.Height = myOPENCV_runlist[i, 2];
sigmaX = (double)myOPENCV_runlist[i, 3];
sigmaY = (double)myOPENCV_runlist[i, 4];
Cv2.GaussianBlur(image_out, image_out, size, sigmaX, sigmaY);
break;
}
case MyOPENCV.medianblur: //中值滤波
{
Cv2.MedianBlur(image_in, image_out, myOPENCV_runlist[i, 1]);
break;
}
case MyOPENCV.bilateralfilter: //双边滤波
{
Mat image_out2 = new Mat();
double sigmaColor, sigmaSpace;
sigmaColor = (double)myOPENCV_runlist[i, 2] * 2;
sigmaSpace = (double)myOPENCV_runlist[i, 3] / 2;
Cv2.BilateralFilter(image_out, image_out2, myOPENCV_runlist[i, 1], sigmaColor, sigmaSpace);
image_out = image_out2;
break;
}
case MyOPENCV.dilate: //膨胀
{
Mat image_element = new Mat();
OpenCvSharp.Size size;
size.Width = myOPENCV_runlist[i, 2];
size.Height = myOPENCV_runlist[i, 3];
image_element = Cv2.GetStructuringElement((MorphShapes)myOPENCV_runlist[i, 1], size);
Cv2.Dilate(image_out, image_out, image_element);
break;
}
case MyOPENCV.erode: //腐蚀
{
Mat image_element = new Mat();
OpenCvSharp.Size size;
size.Width = myOPENCV_runlist[i, 2];
size.Height = myOPENCV_runlist[i, 3];
image_element = Cv2.GetStructuringElement((MorphShapes)myOPENCV_runlist[i, 1], size);
Cv2.Erode(image_out, image_out, image_element);
break;
}
case MyOPENCV.morphologyex: //高级形态学变换
{
Mat image_element = new Mat();
OpenCvSharp.Size size;
size.Width = myOPENCV_runlist[i, 3];
size.Height = myOPENCV_runlist[i, 4];
image_element = Cv2.GetStructuringElement((MorphShapes)myOPENCV_runlist[i, 2], size);
Cv2.MorphologyEx(image_out, image_out, (MorphTypes)myOPENCV_runlist[i, 1], image_element);
break;
}
case MyOPENCV.floodfill: //漫水填充
{
OpenCvSharp.Point point;
point.X = myOPENCV_runlist[i, 1];
point.Y = myOPENCV_runlist[i, 2];
OpenCvSharp.Scalar scalar;
scalar = myOPENCV_runlist[i, 3];
Cv2.FloodFill(image_out, point, scalar);
break;
}
case MyOPENCV.pyrup: //尺寸放大
{
OpenCvSharp.Size size;
size.Width = image_out.Cols * 2;
size.Height = image_out.Rows * 2;
Cv2.PyrUp(image_out, image_out, size);
break;
}
case MyOPENCV.pyrdown: //尺寸缩小
{
OpenCvSharp.Size size;
size.Width = image_out.Cols / 2;
size.Height = image_out.Rows / 2;
Cv2.PyrDown(image_out, image_out, size);
break;
}
case MyOPENCV.resize: //尺寸调整
{
OpenCvSharp.Size size;
InterpolationFlags interpolationFlags;
size.Width = image_out.Cols * myOPENCV_runlist[i, 1] / 10;
size.Height = image_out.Rows * myOPENCV_runlist[i, 2] / 10;
interpolationFlags = (InterpolationFlags)myOPENCV_runlist[i, 3];
Cv2.Resize(image_out, image_out, size, 0, 0, interpolationFlags);
break;
}
case MyOPENCV.threshold: //固定阈值化
{
Cv2.Threshold(image_out, image_out, myOPENCV_runlist[i, 1], myOPENCV_runlist[i, 2], (ThresholdTypes)myOPENCV_runlist[i, 3]);
break;
}
case MyOPENCV.canny: //边缘检测CANNY
{
Mat image_out2 = new Mat();
Cv2.Canny(image_out, image_out2, myOPENCV_runlist[i, 1], myOPENCV_runlist[i, 2], myOPENCV_runlist[i, 3]);
image_out = image_out2;
break;
}
case MyOPENCV.sobel: //边缘检测SOBEL
{
Cv2.Sobel(image_out, image_out, -1, myOPENCV_runlist[i, 1], myOPENCV_runlist[i, 2], myOPENCV_runlist[i, 3]);
break;
}
case MyOPENCV.laplacian: //边缘检测LAPLACIAN
{
myOPENCV_runlist[i, 1] = 0;
Cv2.Laplacian(image_out, image_out, 0, myOPENCV_runlist[i, 2], myOPENCV_runlist[i, 3]);
break;
}
case MyOPENCV.scharr: //边缘检测SCHARR
{
Cv2.Scharr(image_out, image_out, -1, myOPENCV_runlist[i, 1], myOPENCV_runlist[i, 2]);
break;
}
case MyOPENCV.convertscaleabs: //图像快速增强
{
double alpha, beta;
alpha = (double)myOPENCV_runlist[i, 1] / 10;
beta = (double)myOPENCV_runlist[i, 2] / 10;
Cv2.ConvertScaleAbs(image_out, image_out, alpha, beta);
break;
}
case MyOPENCV.addweighted: //图像融合
{
Mat image_in2 = new Mat(my_imagesource2);
double alpha, beta, gamma;
alpha = (double)myOPENCV_runlist[i, 1] / 10;
beta = (double)myOPENCV_runlist[i, 2] / 10;
gamma = (double)myOPENCV_runlist[i, 3] / 10;
Cv2.AddWeighted(image_out, alpha, image_in2, beta, gamma, image_out);
break;
}
case MyOPENCV.houghlines: //霍夫标准变换
{
Scalar scalar = new Scalar(0x00, 0xFF, 0x00); //绿色
LineSegmentPolar[] lines;
OpenCvSharp.Size size = new OpenCvSharp.Size(image_out.Width, image_out.Height);
Mat image_out3 = new Mat(size, MatType.CV_8UC3);
lines = Cv2.HoughLines(image_out, 1, Cv2.PI / 180, myOPENCV_runlist[i, 1]);
for (int ii = 0; ii < lines.Length; ii++)
{
//double rho, theta;
OpenCvSharp.Point pt1, pt2;
double a = Math.Cos(lines[ii].Theta), b = Math.Sin(lines[ii].Theta);
double x0 = a * lines[ii].Rho, y0 = b * lines[ii].Rho;
pt1.X = (int)Math.Round(x0 + 1000 * (-b));
pt1.Y = (int)Math.Round(y0 + 1000 * (a));
pt2.X = (int)Math.Round(x0 - 1000 * (-b));
pt2.Y = (int)Math.Round(y0 - 1000 * (a));
Cv2.Line(image_out3, pt1, pt2, scalar, 1, LineTypes.AntiAlias);
}
if (myOPENCV_runlist[i, 2] == 0)
{
Cv2.AddWeighted(image_out3, (double)myOPENCV_runlist[i, 3] / 10, image_in, (double)myOPENCV_runlist[i, 4] / 10, 0, image_out);
}
else
{
image_out = image_out3;
}
break;
}
case MyOPENCV.houghlinep: //霍夫累计概率变换
{
Scalar scalar = new Scalar(0x00, 0xFF, 0x00); //绿色
LineSegmentPoint[] lines; // 线段检索
OpenCvSharp.Size size = new OpenCvSharp.Size(image_out.Width, image_out.Height);
Mat image_out3 = new Mat(size, MatType.CV_8UC3);
lines = Cv2.HoughLinesP(image_out, 1, Cv2.PI / 180, myOPENCV_runlist[i, 1], myOPENCV_runlist[i, 3], myOPENCV_runlist[i, 4]);
for (int ii = 0; ii < lines.Length; ii++)
{
OpenCvSharp.Point point1, point2;
point1.X = lines[i].P1.X;
point1.Y = lines[i].P1.Y;
point2.X = lines[i].P2.X;
point2.Y = lines[i].P2.Y;
Cv2.Line(image_out3, point1, point2, scalar, 1, LineTypes.AntiAlias);
}
if (myOPENCV_runlist[i, 2] == 0)
{
Cv2.AddWeighted(image_out3, 1, image_in, 0.8, 0, image_out);
}
else
{
image_out = image_out3;
}
break;
}
case MyOPENCV.houghcircles: //霍夫圆变换
{
Scalar scalar = new Scalar(0x00, 0xFF, 0x00); //绿色
CircleSegment[] circles;
OpenCvSharp.Size size = new OpenCvSharp.Size(image_out.Width, image_out.Height);
Mat image_out3 = new Mat(size, MatType.CV_8UC3);
circles = Cv2.HoughCircles(image_out, HoughMethods.Gradient, 1, myOPENCV_runlist[i, 1], myOPENCV_runlist[i, 2], myOPENCV_runlist[i, 3], 0, myOPENCV_runlist[i, 4]);
for (int ii = 0; ii < circles.Length; ii++)
{
OpenCvSharp.Point center;
center.X = (int)Math.Round(circles[ii].Center.X);
center.Y = (int)Math.Round(circles[ii].Center.Y);
int radius = (int)Math.Round(circles[ii].Radius);
Cv2.Circle(image_out3, center.X, center.Y, radius, scalar);
Cv2.Circle(image_out3, center, radius, scalar);
}
Cv2.AddWeighted(image_out3, 1, image_in, 0.6, 0, image_out);
break;
}
case MyOPENCV.remap: //重映射
{
OpenCvSharp.Size size = new OpenCvSharp.Size(image_out.Width, image_out.Height);
Mat map_x = new Mat(size, MatType.CV_32FC1), map_y = new Mat(size, MatType.CV_32FC1);
for (int ii = 0; ii < image_out.Rows; ii++)
{
for (int jj = 0; jj < image_out.Cols; jj++)
{
if (myOPENCV_runlist[i, 1] == 0)
{
map_x.Set <float>(ii, jj, jj); //上下翻转
map_y.Set <float>(ii, jj, image_out.Rows - ii); //上下翻转
}
else if (myOPENCV_runlist[i, 1] == 1)
{
map_x.Set <float>(ii, jj, image_out.Cols - jj); //左右翻转
map_y.Set <float>(ii, jj, ii); //左右翻转
}
else if (myOPENCV_runlist[i, 1] == 2)
{
map_x.Set <float>(ii, jj, image_out.Cols - jj); //上下左右翻转
map_y.Set <float>(ii, jj, image_out.Rows - ii); //上下左右翻转
}
else if (myOPENCV_runlist[i, 1] == 3)
{
map_x.Set <float>(ii, jj, (float)myOPENCV_runlist[i, 2] / 10 * jj); //放大缩小
map_y.Set <float>(ii, jj, (float)myOPENCV_runlist[i, 2] / 10 * ii); //放大缩小
}
}
}
Cv2.Remap(image_out, image_out, map_x, map_y);
break;
}
case MyOPENCV.warpaffine: //仿射变换
{
if (0 == myOPENCV_runlist[i, 1])
{
Mat rot_mat = new Mat(2, 3, MatType.CV_32FC1);
OpenCvSharp.Point center = new OpenCvSharp.Point(image_out.Cols / 2, image_out.Rows / 2);
double angle = myOPENCV_runlist[i, 2];
double scale = (double)myOPENCV_runlist[i, 3] / 10;
///// 通过上面的旋转细节信息求得旋转矩阵
rot_mat = Cv2.GetRotationMatrix2D(center, angle, scale);
///// 旋转已扭曲图像
Cv2.WarpAffine(image_out, image_out, rot_mat, image_out.Size());
}
else
{
Point2f[] srcTri = new Point2f[3];
Point2f[] dstTri = new Point2f[3];
Mat warp_mat = new Mat(2, 3, MatType.CV_32FC1);
Mat warp_dst;
warp_dst = Mat.Zeros(image_out.Rows, image_out.Cols, image_out.Type());
srcTri[0] = new Point2f(0, 0);
srcTri[1] = new Point2f(image_out.Cols, 0);
srcTri[2] = new Point2f(0, image_out.Rows);
dstTri[0] = new Point2f((float)(image_out.Cols * myOPENCV_runlist[i, 2] / 100), (float)(image_out.Rows * myOPENCV_runlist[i, 2] / 100));
dstTri[1] = new Point2f((float)(image_out.Cols * (1 - (float)myOPENCV_runlist[i, 3] / 100)), (float)(image_out.Rows * myOPENCV_runlist[i, 3] / 100));
dstTri[2] = new Point2f((float)(image_out.Cols * myOPENCV_runlist[i, 4] / 100), (float)(image_out.Rows * (1 - (float)myOPENCV_runlist[i, 4] / 100)));
warp_mat = Cv2.GetAffineTransform(srcTri, dstTri);
Cv2.WarpAffine(image_out, image_out, warp_mat, image_out.Size());
}
break;
}
case MyOPENCV.equalizehist: //直方图均衡化
{
Cv2.EqualizeHist(image_out, image_out);
break;
}
case MyOPENCV.facedetection: //人脸识别
{
if (0 == myOPENCV_runlist[i, 1]) // 参数一为0 调用haar，其余数字调用lbp
{
var haarCascade = new CascadeClassifier(@"haarcascade_frontalface_alt.xml");
Mat haarResult = DetectFace(image_out, haarCascade);
image_out = haarResult;
}
else
{
var lbpCascade = new CascadeClassifier(@"lbpcascade_frontalface.xml");
Mat lbpResult = DetectFace(image_out, lbpCascade);
image_out = lbpResult;
}
break;
}
case MyOPENCV.matchtemplate: // 模板匹配
{
Mat originalMat = Cv2.ImRead(my_imagesource, ImreadModes.AnyColor); //母图
Mat modelMat = Cv2.ImRead(my_imagesource2, ImreadModes.AnyColor); //模板
Mat resultMat = new Mat(); // 匹配结果
//resultMat.Create(mat1.Cols - modelMat.Cols + 1, mat1.Rows - modelMat.Cols + 1, MatType.CV_32FC1);//创建result的模板，就是MatchTemplate里的第三个参数
if (0 == myOPENCV_runlist[i, 1])
{
Cv2.MatchTemplate(originalMat, modelMat, resultMat, TemplateMatchModes.SqDiff); //进行匹配(1母图,2模版子图,3返回的result，4匹配模式)
}
else if (1 == myOPENCV_runlist[i, 1])
{
Cv2.MatchTemplate(originalMat, modelMat, resultMat, TemplateMatchModes.SqDiffNormed);
}
else if (2 == myOPENCV_runlist[i, 1])
{
Cv2.MatchTemplate(originalMat, modelMat, resultMat, TemplateMatchModes.CCorr);
}
else if (3 == myOPENCV_runlist[i, 1])
{
Cv2.MatchTemplate(originalMat, modelMat, resultMat, TemplateMatchModes.CCorrNormed);
}
else if (4 == myOPENCV_runlist[i, 1])
{
Cv2.MatchTemplate(originalMat, modelMat, resultMat, TemplateMatchModes.CCoeff);
}
else if (5 == myOPENCV_runlist[i, 1])
{
Cv2.MatchTemplate(originalMat, modelMat, resultMat, TemplateMatchModes.CCoeffNormed);
}
OpenCvSharp.Point minLocation, maxLocation, matchLocation;
Cv2.MinMaxLoc(resultMat, out minLocation, out maxLocation);
matchLocation = maxLocation;
Mat mask = originalMat.Clone();
Cv2.Rectangle(mask, minLocation, new OpenCvSharp.Point(minLocation.X + modelMat.Cols, minLocation.Y + modelMat.Rows), Scalar.Green, 2); //画出匹配的矩 （图像，最小点，最大点，颜色，线宽）
image_out = mask;
break;
}
case MyOPENCV.find_draw_contours: // 找出并绘制轮廓
{
Cv2.CvtColor(image_out, image_out, ColorConversionCodes.RGB2GRAY); //转换为灰度图
//Cv2.Blur(image_out, image_out, new OpenCvSharp.Size(2, 2)); //滤波
Cv2.Canny(image_out, image_out, 100, 200); //Canny边缘检测
OpenCvSharp.Point[][] contours;
HierarchyIndex[] hierarchly;
Cv2.FindContours(image_out, out contours, out hierarchly, RetrievalModes.Tree, ContourApproximationModes.ApproxSimple, new OpenCvSharp.Point(0, 0)); //获得轮廓
Mat dst_Image = Mat.Zeros(image_out.Size(), image_out.Type()); // 图片像素值归零
Random rnd = new Random();
for (int j = 0; j < contours.Length; j++)
{
Scalar color = new Scalar(myOPENCV_runlist[i, 1], myOPENCV_runlist[i, 2], myOPENCV_runlist[i, 3]);
//Scalar color = new Scalar(rnd.Next(0, 255), rnd.Next(0, 255), rnd.Next(0, 255));
Cv2.DrawContours(dst_Image, contours, j, color, myOPENCV_runlist[i, 4], LineTypes.Link8, hierarchly); //画出轮廓
}
image_out = dst_Image;
break;
}
case MyOPENCV.componentdefectdetecting: // 零件缺陷检测
{
Cv2.CvtColor(image_out, image_out, ColorConversionCodes.RGB2GRAY); //转换为灰度图
//Cv2.Blur(image_out, image_out, new OpenCvSharp.Size(2, 2)); //滤波
Cv2.Canny(image_out, image_out, 100, 200); //Canny边缘检测
OpenCvSharp.Point[][] contours;
HierarchyIndex[] hierarchly;
Cv2.FindContours(image_out, out contours, out hierarchly, RetrievalModes.Tree, ContourApproximationModes.ApproxSimple, new OpenCvSharp.Point(0, 0)); //获得轮廓
Mat dst_Image = Mat.Zeros(image_out.Size(), image_out.Type()); // 图片像素值归零
Random rnd = new Random();
for (int j = 0; j < contours.Length; j++)
{
Scalar color = new Scalar(myOPENCV_runlist[i, 1], myOPENCV_runlist[i, 2], myOPENCV_runlist[i, 3]);
//Scalar color = new Scalar(rnd.Next(0, 255), rnd.Next(0, 255), rnd.Next(0, 255));
Cv2.DrawContours(dst_Image, contours, j, color, myOPENCV_runlist[i, 4], LineTypes.Link8, hierarchly); //画出轮廓
}
Mat cnt = new Mat();
Cv2.ConvexHull(image_out, cnt);
break;
}
default: break;
}
}
return(image_out);
}
// Simple example of canny edge detect
void ProcessImage(Mat _image)
{
Cv2.Flip(_image, _image, FlipMode.X);
Cv2.Canny(_image, cannyImage, 100, 100);
var grayImage = new Mat();
Cv2.CvtColor(_image, grayImage, ColorConversionCodes.BGR2GRAY);
Cv2.EqualizeHist(grayImage, grayImage);
var face_cascade = new CascadeClassifier();
face_cascade.Load(Application.dataPath + "/Plugins/Classifiers/haarcascade_frontalface_default.xml");
var eye_cascade = new CascadeClassifier();
eye_cascade.Load(Application.dataPath + "/Plugins/Classifiers/haarcascade_eye_tree_eyeglasses.xml");
var mouth_cascade = new CascadeClassifier();
mouth_cascade.Load(Application.dataPath + "/Plugins/Classifiers/haarcascade_mcs_mouth.xml");
//Debug.Log(" ");
var faces = face_cascade.DetectMultiScale(
image: grayImage,
scaleFactor: 1.3,
minNeighbors: 5,
flags: HaarDetectionType.DoRoughSearch | HaarDetectionType.ScaleImage,
minSize: new Size(100, 100)
);
//Bounds meshRendererBounds = GetComponentInChildren<MeshRenderer>().bounds;
//Vector3 meshRendererCenter = meshRendererBounds.center;
//Vector3 maxBound = meshRendererBounds.max;
//Vector3 minBound = meshRendererBounds.min;
//OpenCvSharp.Rect rect = new OpenCvSharp.Rect((int)meshRendererCenter.x + 350,(int)meshRendererCenter.y + 50, 600,600);
var global_rectangle_color = Scalar.FromRgb(0, 0, 255);
//Cv2.Rectangle(_image, rect, global_rectangle_color, 3);
//Console.WriteLine("Detected faces: {0}", faces.Length);
//Debug.Log(faces.Length);
//var rnd = new System.Random();
var face_count = 0;
var mouth_count = 0;
var eye_count = 0;
foreach (var faceRect in faces)
{
var detectedFaceImage = new Mat(_image, faceRect);
//Cv2.ImShow(string.Format("Face {0}", face_count), detectedFaceImage);
//Cv2.WaitKey(1); // do events
var facec_rectangle_color = Scalar.FromRgb(255, 0, 0);
face = faceRect;
Cv2.Rectangle(_image, faceRect, facec_rectangle_color, 3);
rectFront = new OpenCvSharp.Rect(faceRect.X + faceRect.Width / 2 - 50, faceRect.Y + 50, 100, 50);
//Cv2.Rectangle(_image, rectFront, global_rectangle_color, 3);
var detectedFaceGrayImage = new Mat();
Cv2.CvtColor(detectedFaceImage, detectedFaceGrayImage, ColorConversionCodes.BGRA2GRAY);
var eyes = eye_cascade.DetectMultiScale(
image: grayImage,
scaleFactor: 1.3,
minNeighbors: 5,
flags: HaarDetectionType.DoRoughSearch | HaarDetectionType.ScaleImage,
minSize: new Size(50, 50)
);
foreach (var eyeRect in eyes)
{
var detectedEyeImage = new Mat(_image, eyeRect);
//Cv2.ImShow(string.Format("Face {0}", eye_count), detectedEyeImage);
//Cv2.WaitKey(1); // do events
var eye_rectangle_color = Scalar.FromRgb(0, 255, 0);
//Cv2.Rectangle(_image, eyeRect, eye_rectangle_color, 3);
if (eye_count == 1)
{
// Par rapport à la position de l'oeil gauche
rectEyeLeft = new OpenCvSharp.Rect(eyeRect.X + 75, eyeRect.Y + 100, 25, 25);
//Cv2.Rectangle(_image, rectEyeLeft, global_rectangle_color, 3);
}
else
{
// Par rapport à la position de l'oeil droit
rectEyeRight = new OpenCvSharp.Rect(eyeRect.X, eyeRect.Y + 100, 25, 25);
//Cv2.Rectangle(_image, rectEyeRight, global_rectangle_color, 3);
}
var detectedEyeGrayImage = new Mat();
Cv2.CvtColor(detectedEyeImage, detectedEyeGrayImage, ColorConversionCodes.BGRA2GRAY);
eye_count++;
}
var mouth = mouth_cascade.DetectMultiScale(
image: grayImage,
scaleFactor: 1.3,
minNeighbors: 5,
flags: HaarDetectionType.DoRoughSearch | HaarDetectionType.ScaleImage,
minSize: new Size(50, 50)
);
foreach (var m in mouth)
{
var detectedEarImage = new Mat(_image, m);
//Cv2.ImShow(string.Format("Face {0}", eye_count), detectedEyeImage);
//Cv2.WaitKey(1); // do events
if (m.Y > eyes[0].Y && Mathf.Abs(m.Y - eyes[0].Y) > 100)
{
//Debug.Log("mouth height :");
//Debug.Log(m.Height);
var eye_rectangle_color = Scalar.FromRgb(0, 255, 0);
Cv2.Rectangle(_image, m, eye_rectangle_color, 3);
lipHeight = (float)m.Height / (float)face.Height;
}
var detectedEyeGrayImage = new Mat();
Cv2.CvtColor(detectedEarImage, detectedEyeGrayImage, ColorConversionCodes.BGRA2GRAY);
mouth_count++;
}
face_count++;
}
}
private static void SegmentationCannyFilledPolygons(Camera camera, out List <Point2f> v_center, out List <float> v_radius)
{
RenderTexture activeRenderTexture = RenderTexture.active;
RenderTexture.active = camera.targetTexture;
camera.Render();
Texture2D currentFrame = new Texture2D(camera.targetTexture.width, camera.targetTexture.height);
currentFrame.ReadPixels(new UnityEngine.Rect(0, 0, camera.targetTexture.width, camera.targetTexture.height), 0, 0);
currentFrame.Apply();
RenderTexture.active = activeRenderTexture;
Mat image = OpenCvSharp.Unity.TextureToMat(currentFrame);
UnityEngine.Object.Destroy(currentFrame);
Mat grayImage = new Mat();
Cv2.CvtColor(image, grayImage, ColorConversionCodes.BGR2GRAY);
Cv2.EqualizeHist(grayImage, grayImage);
Cv2.GaussianBlur(grayImage, grayImage, new Size(9, 9), 2, 2);
Mat edgesImage = new Mat();
Cv2.Canny(grayImage, edgesImage, 40, 20);
Point[][] contours_canny;
HierarchyIndex[] hierarchy_canny;
Cv2.FindContours(edgesImage, out contours_canny, out hierarchy_canny, RetrievalModes.List, ContourApproximationModes.ApproxSimple, null);
Mat img_all_contours_and_filled = Mat.Zeros(edgesImage.Height, edgesImage.Width, MatType.CV_8UC1);
Mat img_all_contours = Mat.Zeros(edgesImage.Height, edgesImage.Width, MatType.CV_8UC1);
for (int j = 0; j < contours_canny.Length; j++)
{
Scalar color = new Scalar(255, 255, 255);
Cv2.DrawContours(img_all_contours_and_filled, contours_canny, j, color, -1, LineTypes.Link8, hierarchy_canny);
Cv2.DrawContours(img_all_contours, contours_canny, j, color, 1, LineTypes.Link8, hierarchy_canny);
}
Mat img_only_closed_contours = new Mat();
Cv2.Absdiff(img_all_contours_and_filled, img_all_contours, img_only_closed_contours);
Point[][] contours;
HierarchyIndex[] hierarchy;
Cv2.FindContours(img_only_closed_contours, out contours, out hierarchy, RetrievalModes.External, ContourApproximationModes.ApproxTC89L1, null);
Point[][] contours_poly = new Point[contours.Length][];
OpenCvSharp.Rect[] boundRect = new OpenCvSharp.Rect[contours.Length];
List <Point2f> contours_center = new List <Point2f> {
};
List <float> contours_radius = new List <float> {
};
int i_contour = 0;
foreach (Point[] contour in contours)
{
Point2f contour_center;
float contour_radius;
contours_poly[i_contour] = Cv2.ApproxPolyDP(contour, 3, true);
Cv2.MinEnclosingCircle(contours_poly[i_contour], out contour_center, out contour_radius);
//currentFrame = DrawCircle(currentFrame, (int)contour_center.X, (int)contour_center.Y, (int)contour_radius);
contours_center.Add(contour_center);
contours_radius.Add(contour_radius);
i_contour++;
}
v_center = contours_center;
v_radius = contours_radius;
//TextureToPNG(currentFrame);
}
void Update()
{
// If the webcam has been updated, update the frame.
if (capture.Read(frame))
{
// Convert the frame into grayscale.
Cv2.CvtColor(frame, gray, ColorConversionCodes.RGB2GRAY);
// Equalise lighting to make landmark detection more accurate. - detecting points.
Cv2.EqualizeHist(gray, gray);
}
//// As faces is easiest to detect, check for these first.
OpenCvSharp.Rect[] faces = new OpenCvSharp.Rect[0];
faces = haarCascade.DetectMultiScale(
gray, 1.15, 5, HaarDetectionType.DoRoughSearch | HaarDetectionType.ScaleImage, new Size(60, 60));
//// Detect if any eyes are in the image.
//OpenCvSharp.Rect[] eyes = new OpenCvSharp.Rect[2];
//eyes = eyeCascade.DetectMultiScale(
// gray, 1.15, 6, HaarDetectionType.DoRoughSearch | HaarDetectionType.ScaleImage, new Size(40, 40));
//// If eyes are equal = 2 then there must be a face. - assumption is that the two eyes are from the same person.
//if (eyes.Length == 2 && faces.Length == 0)
//{
// Point averagePoint = new Point();
// averagePoint.X = (eyes[0].X + eyes[1].X) / 2;
// averagePoint.Y = (eyes[0].Y + eyes[1].Y) / 2;
// faces = new OpenCvSharp.Rect[1];
// faces[0] = new OpenCvSharp.Rect(averagePoint, new Size(150, 150));
//}
//// If there is an eye, there must be a face. Not as accurate as there is no other landmark to triangulate point.
//else if (eyes.Length == 1 && faces.Length == 0)
//{
// Point averagePoint = new Point();
// averagePoint.X = eyes[0].X;
// averagePoint.Y = eyes[0].Y - 2;
// faces = new OpenCvSharp.Rect[1];
// faces[0] = new OpenCvSharp.Rect(averagePoint, new Size(80, 80));
//}
//// due to cascade method not being rotation invarient, rotate frame and check if face is there.
//Mat clockRot = new Mat();
//// If no faces were detected, rotate image.
//if (faces.Length == 0)
//{
// clockRot = RotateFrame(gray, 40);
// faces = haarCascade.DetectMultiScale(
// clockRot, 1.15, 5, HaarDetectionType.DoRoughSearch | HaarDetectionType.ScaleImage, new Size(60, 60));
//}
//if (faces.Length == 0)
//{
// clockRot = RotateFrame(gray, -40);
// faces = haarCascade.DetectMultiScale(
// clockRot, 1.15, 5, HaarDetectionType.DoRoughSearch | HaarDetectionType.ScaleImage, new Size(60, 60));
//}
//// For each face detected, add square around feature, and check for eyes.
foreach (var faceRect in faces)
{
// Obain area face is contained.
var faceMat = new Mat(frame, faceRect);
// Set color of rectangle for face.
var color = Scalar.FromRgb(0, 0, 255);
// Draw rectangle around face. - remove once edge is found.
Cv2.Rectangle(frame, faceRect, color, 3);
}
//foreach (var eyeRect in eyes)
//{
// var eyeMat = new Mat(gray, eyeRect);
// var color = Scalar.FromRgb(0, 255, 0);
// Cv2.Rectangle(frame, eyeRect, color, 3);
//}
// This segment is to draw contours around an image and display it on the OpenCv window.
// Blur the image.
Cv2.Blur(gray, gray, new Size(3, 3));
// Detect edges using canny - Lower number = more contours.
Cv2.Canny(gray, cannyOutput, 100, 255);
// Cv2.Threshold(gray, cannyOutput,100,255,ThresholdTypes.BinaryInv);
// Find contours
Cv2.FindContours(cannyOutput, out contours, out hierarchy, RetrievalModes.Tree, ContourApproximationModes.ApproxTC89L1, new Point(0, 0));
// Draw contours
//Parallel.For(0, contours.Length, i =>
//{
// Cv2.DrawContours(frame, contours, i, Scalar.White, 2, LineTypes.Filled, hierarchy);
//});
if (contours != null)
{
for (int i = 0; i < contours.Length; i++)
{
Cv2.DrawContours(frame, contours, i, Scalar.White, 2, LineTypes.Link8, hierarchy);
if (Input.GetKeyDown("space"))
{
Application.Quit();
}
}
}
//Debug.Log("Drawn contour.");
// Update video.
Cv2.ImShow("Contours", frame);
// frame.SaveImage("image",);
// Cv2.ImWrite("image.jpg", frame);
// snapFrame = Resources.Load("image.jpg") as Texture2D;
//Debug.Log(snapFrame.dimension);
}
public static void RunFacialDetection(CancellationTokenSource cts, FaceDetectionService service)
{
string haarCascadeFile = RuntimeInformation.IsOSPlatform(OSPlatform.Linux) ? _piSettings.LinuxPathToFaceHaarCascade
: _piSettings.DevPathToFaceHaarCascade;
Console.WriteLine("Path to cascade classifier: " + haarCascadeFile);
if (!File.Exists(haarCascadeFile))
{
Console.WriteLine("NO HAAR FILE FOUND");
return;
}
Mat sourceImg = new Mat();
DateTime lastFaceTime = DateTime.Now;
bool wasSearchingForFace = true;
VideoCapture captureInstance = new VideoCapture(0);
while (!captureInstance.IsOpened())
{
Console.WriteLine("Video Capture being reopened.");
captureInstance.Open(0);
Thread.Sleep(500);
}
using (CascadeClassifier cascade = new CascadeClassifier(haarCascadeFile))
//using (Window webCamWindow = new Window("webCamWindow"))
{
while (!cts.IsCancellationRequested)
{
captureInstance.Read(sourceImg);
if (sourceImg.Empty())
{
break;
}
var grayImage = new Mat();
Cv2.CvtColor(sourceImg, grayImage, ColorConversionCodes.BGRA2GRAY);
Cv2.EqualizeHist(grayImage, grayImage);
var faces = cascade.DetectMultiScale(
image: grayImage,
scaleFactor: _faceDetectionConfiguration.ScaleFactor,
minNeighbors: _faceDetectionConfiguration.MinimumNeighbors,
flags: HaarDetectionType.DoRoughSearch | HaarDetectionType.ScaleImage,
minSize: new Size(_faceDetectionConfiguration.MinimumFaceWidth, _faceDetectionConfiguration.MinimumFaceHeight)
);
if (faces.Length > 0)
{
lastFaceTime = DateTime.Now;
if (wasSearchingForFace)
{
service.PublishFaceDetectionStats(new FaceDetectionStats
{
FaceRectangles = faces.Select(face => new Shared.FaceRect
{
X = face.X,
Y = face.Y,
Width = face.Width,
Height = face.Height
}).ToList()
});
wasSearchingForFace = false;
}
}
else if (DateTime.Now - lastFaceTime >= TimeSpan.FromSeconds(_faceDetectionConfiguration.FaceTimeoutInSeconds))
{
if (!wasSearchingForFace)
{
service.PublishFaceDetectionStats(new FaceDetectionStats
{
IsZeroFaceAlert = true
});
}
wasSearchingForFace = true;
}
}
}
}
static void Main(string[] args)
{
var srcImage = new Mat(@"..\..\Images\Test.jpg");
Cv2.ImShow("Source", srcImage);
Cv2.WaitKey(1); // do events
var grayImage = new Mat();
Cv2.CvtColor(srcImage, grayImage, ColorConversion.BgrToGray);
Cv2.EqualizeHist(grayImage, grayImage);
var cascade = new CascadeClassifier(@"..\..\Data\haarcascade_frontalface_alt.xml");
var nestedCascade = new CascadeClassifier(@"..\..\Data\haarcascade_eye_tree_eyeglasses.xml");
var faces = cascade.DetectMultiScale(
image: grayImage,
scaleFactor: 1.1,
minNeighbors: 2,
flags: HaarDetectionType.Zero | HaarDetectionType.ScaleImage,
minSize: new Size(30, 30)
);
Console.WriteLine("Detected faces: {0}", faces.Length);
var rnd = new Random();
var count = 1;
foreach (var faceRect in faces)
{
var detectedFaceImage = new Mat(srcImage, faceRect);
Cv2.ImShow(string.Format("Face {0}", count), detectedFaceImage);
Cv2.WaitKey(1); // do events
var color = Scalar.FromRgb(rnd.Next(0, 255), rnd.Next(0, 255), rnd.Next(0, 255));
Cv2.Rectangle(srcImage, faceRect, color, 3);
var detectedFaceGrayImage = new Mat();
Cv2.CvtColor(detectedFaceImage, detectedFaceGrayImage, ColorConversion.BgrToGray);
var nestedObjects = nestedCascade.DetectMultiScale(
image: detectedFaceGrayImage,
scaleFactor: 1.1,
minNeighbors: 2,
flags: HaarDetectionType.Zero | HaarDetectionType.ScaleImage,
minSize: new Size(30, 30)
);
Console.WriteLine("Nested Objects[{0}]: {1}", count, nestedObjects.Length);
foreach (var nestedObject in nestedObjects)
{
var center = new Point
{
X = Cv.Round(nestedObject.X + nestedObject.Width * 0.5) + faceRect.Left,
Y = Cv.Round(nestedObject.Y + nestedObject.Height * 0.5) + faceRect.Top
};
var radius = Cv.Round((nestedObject.Width + nestedObject.Height) * 0.25);
Cv2.Circle(srcImage, center, radius, color, thickness: 3);
}
count++;
}
Cv2.ImShow("Haar Detection", srcImage);
Cv2.WaitKey(1); // do events
Cv2.WaitKey(0);
Cv2.DestroyAllWindows();
srcImage.Dispose();
}
/// <summary>
/// Detector
/// </summary>
/// <param name="inputTexture">Input Unity texture</param>
/// <param name="texParams">Texture parameters (flipped, rotated etc.)</param>
/// <param name="detect">Flag signalling whether we need detection on this frame</param>
public virtual void ProcessTexture(T texture, Unity.TextureConversionParams texParams, bool detect = true)
{
// convert Unity texture to OpenCv::Mat
ImportTexture(texture, texParams);
// detect
if (detect)
{
double invF = 1.0 / appliedFactor;
DataStabilizer.ThresholdFactor = invF;
// convert to grayscale and normalize
Mat gray = new Mat();
Cv2.CvtColor(processingImage, gray, ColorConversionCodes.BGR2GRAY);
// fix shadows
Cv2.EqualizeHist(gray, gray);
/*Mat normalized = new Mat();
* CLAHE clahe = CLAHE.Create();
* clahe.TilesGridSize = new Size(8, 8);
* clahe.Apply(gray, normalized);
* gray = normalized;*/
// detect matching regions (faces bounding)
Rect[] rawFaces = cascadeFaces.DetectMultiScale(gray, 1.2, 6);
if (Faces.Count != rawFaces.Length)
{
Faces.Clear();
}
// now per each detected face draw a marker and detect eyes inside the face rect
int facesCount = 0;
for (int i = 0; i < rawFaces.Length; ++i)
{
Rect faceRect = rawFaces[i];
Rect faceRectScaled = faceRect * invF;
using (Mat grayFace = new Mat(gray, faceRect))
{
// another trick: confirm the face with eye detector, will cut some false positives
if (cutFalsePositivesWithEyesSearch && null != cascadeEyes)
{
Rect[] eyes = cascadeEyes.DetectMultiScale(grayFace);
if (eyes.Length == 0 || eyes.Length > 2)
{
continue;
}
}
// get face object
DetectedFace face = null;
if (Faces.Count < i + 1)
{
face = new DetectedFace(DataStabilizer, faceRectScaled);
Faces.Add(face);
}
else
{
face = Faces[i];
face.SetRegion(faceRectScaled);
}
// shape
facesCount++;
if (null != shapeFaces)
{
Point[] marks = shapeFaces.DetectLandmarks(gray, faceRect);
// we have 68-point predictor
if (marks.Length == 68)
{
// transform landmarks to the original image space
List <Point> converted = new List <Point>();
foreach (Point pt in marks)
{
converted.Add(pt * invF);
}
// save and parse landmarks
face.SetLandmarks(converted.ToArray());
}
}
}
}
// log
//UnityEngine.Debug.Log(String.Format("Found {0} faces", Faces.Count));
}
}
//车牌颜色定位
public static List <Mat> PlateLocateByColor(Mat matIn, ParameterList pl)
{
List <Mat> matPlates = new List <Mat>();
Mat matHsv = matIn.CvtColor(ColorConversionCodes.BGR2HSV);
//对v均衡化后在合并
Mat[] matToHsv = new Mat[3];
Cv2.Split(matHsv, out matToHsv);
Cv2.EqualizeHist(matToHsv[2], matToHsv[2]);
Mat equalizeHistHsv = new Mat();
Cv2.Merge(matToHsv, equalizeHistHsv);
//在均衡化后的hsv颜色空间红寻找黄色和蓝色区域
Mat matYellow = new Mat();
Mat matBlue = new Mat();
Scalar yellow_low = new Scalar(15, 70, 70);
Scalar yellow_up = new Scalar(40, 255, 255);
Scalar blue_low = new Scalar(100, 70, 70);
Scalar blue_up = new Scalar(140, 255, 255);
Cv2.InRange(equalizeHistHsv, yellow_low, yellow_up, matYellow);
Cv2.InRange(equalizeHistHsv, blue_low, blue_up, matBlue);
Mat matAll = matBlue + matYellow;
//使用形态学操作对选定颜色区域进行处理
Mat matAllDilate = new Mat();
Mat matAllErode = new Mat();
Mat element = Cv2.GetStructuringElement(MorphShapes.Rect, new OpenCvSharp.Size(7, 3));
Cv2.Dilate(matAll, matAllDilate, element);
Cv2.Erode(matAllDilate, matAllErode, element);
//寻找轮廓
OpenCvSharp.Point[][] contours; //vector<vector<Point>> contours;
HierarchyIndex[] hierarchyIndexes; //vector<Vec4i> hierarchy;
Cv2.FindContours(
matAllErode,
out contours,
out hierarchyIndexes,
mode: RetrievalModes.Tree,
method: ContourApproximationModes.ApproxSimple); //求轮廓
//求轮廓的最小外接矩形
List <Rect> rects = new List <Rect>();
foreach (OpenCvSharp.Point[] p in contours)
{
Rect rect = Cv2.BoundingRect(p);
if ((double)rect.Height / rect.Width > pl.HeightDivideWidthLow && (double)rect.Height / rect.Width < pl.HeightDivedeWidthUp &&
rect.Height > pl.HeightLow && rect.Height < pl.HeightUp &&
rect.Width > pl.WidthLow && rect.Width < pl.WidthUp)
{
rects.Add(rect);
}
}
if (rects.Count == 0)
{
return(matPlates);
}
for (int index = 0; index < rects.Count; index++)
{
Mat roi = new Mat(matIn, rects[index]);
matPlates.Add(roi);
}
return(matPlates);
}
public void Execute(TextDlg text)
{
Output = Input.Clone();
InputBuffer.Add(Input);
BufferState++;
BufferState %= BufferSize;
TimesBuffer.Add(DateTime.Now - StartTime);
if (!LockFace)
{
var input_g = new Mat();
Cv2.CvtColor(Input, input_g, ColorConversionCodes.BGR2GRAY);
Cv2.EqualizeHist(input_g, input_g);
var detect = Classifier.DetectMultiScale(input_g, minNeighbors: 4, minSize: new Size(50, 50),
flags: HaarDetectionType.ScaleImage);
if (detect.Length > 0)
{
var biggest = detect.OrderByDescending(r => r.Width * r.Height).First();
if (CalcShift(biggest) > SHIFT_THRESHOLD)
{
Face = biggest;
InputBuffer.Clear();
}
}
}
if (Face == OpenCvSharp.Rect.Empty)
{
AverageColorBuffer.Add(0);
return;
}
var forehead = GetSlice(0.5, 0.15, 0.38, 0.14);
Rect(Face, BLUE);
Rect(forehead, GREEN);
if (Colorify)
{
// todo
}
var fh_average = CalcMeanColor(forehead);
AverageColorBuffer.Add(fh_average);
var num_samples = TimesBuffer.Count;
if (num_samples > MIN_SAMPLE_COUNT)
{
var time_start = TimesBuffer[0];
var time_end = TimesBuffer[TimesBuffer.Count - 1];
LastFPS = FPS;
var dt = time_end - time_start;
if (dt != TimeSpan.Zero)
{
var fps = num_samples / dt.TotalSeconds;
FPS = (int)(fps * FPS_SMOOTHING + LastFPS * (1 - FPS_SMOOTHING));
}
var linear = MathNet.Numerics.Generate.LinearSpaced(num_samples, time_start.TotalSeconds, time_end.TotalSeconds);
var window = MathNet.Numerics.Window.Hamming(num_samples);
var interp =
window.Zip(linear.Select(i => MathNet.Numerics.Interpolate.Linear(TimesBuffer.Select(x => x.TotalSeconds), AverageColorBuffer).Interpolate(i)), (a, b) => a - b).ToList();
Deviation = interp.Select(x => x - ProgressiveMean(interp)).ToList();
var fourier_raw = Deviation.Select(x => new Complex32((float)x, 0)).ToArray();
MathNet.Numerics.IntegralTransforms.Fourier.Forward(fourier_raw);
Fourier = fourier_raw.Select(x => Math.Abs(x.Real)).ToArray();
Frequencies = Enumerable.Range(0, (int)Math.Ceiling(num_samples / 2d + 1)).Select(x => x * 60.0 * FPS / num_samples).ToArray();
try
{
var zip = Frequencies.Zip(Fourier, (a, b) => new { a, b })
.Where(x => x.a > BPM_LOW && x.b < BPM_HIGH).ToList();
Frequencies = zip.Select(x => x.a).ToArray();
Fourier = zip.Select(x => x.b).ToArray();
BPMPosition = Array.IndexOf(Fourier, Fourier.Max());
BPM = Frequencies[BPMPosition];
BPMBuffer.Add(BPM);
CorrectBPM = ProgressiveMean(BPMBuffer);
text($"d={Deviation.Last():F3}", 20, 120);
}
catch
{
;
}
if (FPS != 0)
{
Gap = (BufferSize - num_samples) / FPS;
}
}
}
private void GrabFace(Mat mat)
{
var srcImage = mat;
_(String.Format("Loading image..."));
Cv2.WaitKey(1);
var grayImage = new Mat();
// _("Generating greyscale");
Cv2.CvtColor(srcImage, grayImage, ColorConversionCodes.BGRA2GRAY);
// _("Creating histogram...");
Cv2.EqualizeHist(grayImage, grayImage);
// _("Loading Haar cascade...");
var cascade = new CascadeClassifier("D:\\cascades\\haarcascade_frontalface_default.xml");
// var nestedCascade = new CascadeClassifier("D:\\cascades\\haarcascade_eye_tree_eyeglasses.xml");
// _("Detecting faces...");
var _faces = cascade.DetectMultiScale(
image: grayImage,
scaleFactor: 1.1,
minNeighbors: 3,
flags: HaarDetectionType.DoCannyPruning | HaarDetectionType.DoRoughSearch | HaarDetectionType.ScaleImage,
minSize: new OpenCvSharp.Size(30, 30)
);
// _(String.Format("Faces detected: {0}", _faces.Length));
if (_faces.Length > 0)
{
_("Face found! Extracting...");
foreach (var faceRect in _faces)
{
var faceBitmap = new Bitmap(mat.ToBitmap());
using (var g = Graphics.FromImage(faceBitmap))
{
g.FillRectangle(new SolidBrush(System.Drawing.Color.FromArgb(200, 0, 0, 0)), g.ClipBounds);
var br = new SolidBrush(System.Drawing.Color.FromArgb(200, System.Drawing.Color.LightGreen));
var box = new System.Drawing.Rectangle(faceRect.X, faceRect.Y, faceRect.Width, faceRect.Height);
g.DrawRectangle(Pens.LightGreen, box);
int rTop = faceRect.Top + faceRect.Height;
g.FillRectangle(br, faceRect.Left, rTop, faceRect.Width, faceRect.Height);
ImgPreviewBox.Source = faceBitmap.ToBitmapSource();
}
count++;
var detectedFaceImage = new Mat(srcImage, faceRect);
if (groupName != "" && personName != "")
{
String iPath = String.Format("{0}{1}\\{2}\\{3}.png", baseDir, groupName, personName, count);
_(String.Format("Saving {0}...", iPath));
imagePaths.Add(iPath);
detectedFaceImage.SaveImage(iPath);
GeneratePreview(iPath);
}
}
}
}
public CameraThread(int id = 0) : base($"camera-thread_{id}")
{
UpdateInterval = 0;
SerialPort.Open();
ThreadStarted += new EventHandler <ThreadService>((obj, service) =>
{
try
{
_capture = new VideoCapture(0);
if (_capture.Read(_mat))
{
_capture.Read(_mat);
if (_mat?.Width == 0)
{
Stop();
MessageBox.Show("Camera Unavailable");
}
}
}
catch
{
Stop();
MessageBox.Show("Camera Unavailable");
}
});
Update += new EventHandler <ThreadService>((obj, service) =>
{
if (_capture.Read(_mat))
{
var grayImage = new Mat();
_mat = _mat.Flip(FlipMode.Y);
Cv2.CvtColor(_mat, grayImage, ColorConversionCodes.BGRA2GRAY);
Cv2.EqualizeHist(grayImage, grayImage);
var faces = _cascade.DetectMultiScale(
image: grayImage,
scaleFactor: 1.6,
minNeighbors: 3,
flags: HaarDetectionType.DoCannyPruning | HaarDetectionType.ScaleImage,
minSize: new OpenCvSharp.Size(30, 30)
);
var biggestFace = faces.FirstOrDefault();
var faceSize = biggestFace.Size.Width * biggestFace.Size.Height;
foreach (var faceRect in faces)
{
if (faceRect.Width * faceRect.Height > biggestFace.Width * biggestFace.Height)
{
biggestFace = faceRect;
faceSize = biggestFace.Width * biggestFace.Height;
}
}
if (faceSize > 0)
{
if (_previousFaceSize == faceSize)
{
_count++;
}
else
{
_count = 0;
}
_previousFaceSize = faceSize;
}
else
{
_count = 0;
}
if (_count > 3)
{
int x = biggestFace.X + biggestFace.Width / 2;
int y = biggestFace.Y + biggestFace.Height / 2 - biggestFace.Height / 4;
int dx = x - _lastX;
int dy = y - _lastY;
_lastX = x;
_lastY = y;
if (Math.Abs(dx) < 50)
{
x += 40 * Math.Sign(dx);
}
if (Math.Abs(dy) < 50)
{
y += 40 * Math.Sign(dy);
}
Cv2.Rectangle(_mat, biggestFace, Scalar.Red, 3);
SerialPort.Write(
$"{getH(x, _previousFaceSize)};" +
$"{getV(y, _previousFaceSize)}");
}
// Turns proccessed camera image into bitmap and passes it to form.
if (!_doCancel)
{
ImageProcessed?.Invoke(this, _mat.ToBitmap());
}
}
// Horizontal linear equation.
// h(x) = ax + c
int getH(int x, int size)
{
double distance = Math.Sqrt(size) - 65;
double a = (0.0726 - (37.899 / 1000_000) * distance);
double c = 75.502 - (12.651 / 1000) * distance;
return((int)(a * x + c));
}
// Vertical linear equation.
// v(y) = ay + c
int getV(int y, int size)
{
double distance = Math.Sqrt(size) - 65;
double a = (0.0714 - (39.343 / 1000_000) * distance);
double c = 61.553 + (81.848 / 1000) * distance;
return((int)(a * y + c));
}
// Useful for calibration to avoid servo sliding
//if (MoveServos)
//{
// SerialPort.Write($"{H - 40};{V - 40}");
// Thread.Sleep(400);
// SerialPort.Write($"{H};{V}");
// MoveServos = false;
//}
});
}
static void Main(string[] args)
{
Mat mat1 = null, mat2 = null, mat3 = null;
var frame = new Mat();
var capture = new VideoCapture(0);
capture.Open(0);
if (capture.IsOpened())
{
while (true)
{
capture.Read(frame);
Console.WriteLine("read");
mat3 = frame;
if (mat1 != null)
{
//var image = DiffImage(mat3, mat2, mat1).ToBitmap();
Mat d1 = new Mat();
Cv2.Absdiff(mat3, mat2, d1);
var image = mat3.ToBitmap();
//image.Save("1.jpg", ImageFormat.Jpeg);
//Console.WriteLine("save");
var grayImage = new Mat();
Cv2.CvtColor(mat3, grayImage, ColorConversionCodes.BGRA2GRAY);
Cv2.EqualizeHist(grayImage, grayImage);
var cascade = new CascadeClassifier(@".\CascadeClassifiers\haarcascade_frontalface_alt2.xml");
var nestedCascade = new CascadeClassifier(@".\CascadeClassifiers\haarcascade_eye_tree_eyeglasses.xml");
var faces = cascade.DetectMultiScale(
image: grayImage,
scaleFactor: 1.1,
minNeighbors: 2,
flags: HaarDetectionType.DoRoughSearch | HaarDetectionType.ScaleImage,
minSize: new Size(30, 30)
);
Console.WriteLine("Detected faces: {0}", faces.Length);
var srcImage = mat3;
//new BodyDetector().Detect(mat3);
faces.ToList().ForEach(e => e.Height += 10);
foreach (var faceRect in faces)
{
Cv2.Rectangle(frame, faceRect, Scalar.Red, 2);
//var a = new Mat(srcImage, faceRect);
//var eigenValues = OutputArray.Create(a);
//var eigenVectors = OutputArray.Create(a);
//Cv2.Eigen(a, eigenValues, eigenVectors);
}
Cv2.ImShow("Source", mat3);
Cv2.WaitKey(1); // do events
//var count = 1;
//foreach (var faceRect in faces)
//{
// var detectedFaceImage = new Mat(srcImage, faceRect);
// Cv2.ImShow(string.Format("Face {0}", count), detectedFaceImage);
// Cv2.WaitKey(1); // do events
// //var color = Scalar.FromRgb(rnd.Next(0, 255), rnd.Next(0, 255), rnd.Next(0, 255));
// Cv2.Rectangle(srcImage, faceRect, Scalar.Red, 3);
// var detectedFaceGrayImage = new Mat();
// Cv2.CvtColor(detectedFaceImage, detectedFaceGrayImage, ColorConversionCodes.BGRA2GRAY);
// var nestedObjects = nestedCascade.DetectMultiScale(
// image: detectedFaceGrayImage,
// scaleFactor: 1.1,
// minNeighbors: 2,
// flags: HaarDetectionType.DoRoughSearch | HaarDetectionType.ScaleImage,
// minSize: new Size(30, 30)
// );
// Console.WriteLine("Nested Objects[{0}]: {1}", count, nestedObjects.Length);
// foreach (var nestedObject in nestedObjects)
// {
// var center = new Point
// {
// X = (int)(Math.Round(nestedObject.X + nestedObject.Width * 0.5, MidpointRounding.ToEven) + faceRect.Left),
// Y = (int)(Math.Round(nestedObject.Y + nestedObject.Height * 0.5, MidpointRounding.ToEven) + faceRect.Top)
// };
// var radius = Math.Round((nestedObject.Width + nestedObject.Height) * 0.25, MidpointRounding.ToEven);
// Cv2.Circle(srcImage, center, (int)radius, Scalar.Red, thickness: 3);
// }
// count++;
//}
//Cv2.ImShow("Haar Detection", srcImage);
//Cv2.WaitKey(1); // do events
}
mat1 = mat2;
mat2 = mat3;
}
}
}
private void DemoFaceTrack()
{
int ColorWidth = kinectManager.ColorWidth;
int ColorHeight = kinectManager.ColorHeight;
Mat colorImage = new Mat(kinectManager.ColorHeight, ColorWidth, MatType.CV_8UC4, kinectManager.ColorRawData); //rows=height, cols=width
Mat grayImage = new Mat();
Cv2.CvtColor(colorImage, grayImage, ColorConversionCodes.RGBA2GRAY);
Cv2.EqualizeHist(grayImage, grayImage);
OpenCvSharp.Rect[] faces = cascade.DetectMultiScale(
image: grayImage,
scaleFactor: 1.1,
minNeighbors: 2,
flags: HaarDetectionType.DoRoughSearch | HaarDetectionType.ScaleImage,
minSize: new Size(30, 30)
);
for (int i = 0; i < faces.Length; i++)
{
OpenCvSharp.Rect faceRect = faces[i];
//outline overall face in image
var rndColor = Scalar.FromRgb(
UnityEngine.Random.Range(0, 255),
UnityEngine.Random.Range(0, 255),
UnityEngine.Random.Range(0, 255)
);
Cv2.Rectangle(colorImage, faceRect, rndColor, 3);
//now do nested features like the eyes
Mat subFaceImage = new Mat(colorImage, faceRect);
Mat detectedFaceGrayImage = new Mat();
Cv2.CvtColor(subFaceImage, detectedFaceGrayImage, ColorConversionCodes.RGBA2GRAY);
var nestedObjects = nestedCascade.DetectMultiScale(
image: detectedFaceGrayImage,
scaleFactor: 1.1,
minNeighbors: 2,
flags: HaarDetectionType.DoRoughSearch | HaarDetectionType.ScaleImage,
minSize: new Size(30, 30)
);
//display each nested face feature
foreach (var nestedObject in nestedObjects)
{
var center = new Point
{
X = (int)(Math.Round(nestedObject.X + nestedObject.Width * 0.5, MidpointRounding.ToEven) + faceRect.Left),
Y = (int)(Math.Round(nestedObject.Y + nestedObject.Height * 0.5, MidpointRounding.ToEven) + faceRect.Top)
};
var radius = Math.Round((nestedObject.Width + nestedObject.Height) * 0.25, MidpointRounding.ToEven);
Cv2.Circle(colorImage, center, (int)radius, rndColor, thickness: 3);
}
}
//Convert back to RGBA32
//Mat faceImageOut = new Mat(IRWidth, IRHeight, MatType.CV_8UC4);
//Cv2.CvtColor(colorImage, faceImageOut, ColorConversionCodes.RGBA2RGBA);
//load onto texture
byte[] rawTextureBytes = KinectCVUtilities.ConvertMatToBytes(colorImage);
kinectManager.ColorTexture.LoadRawTextureData(rawTextureBytes);
kinectManager.ColorTexture.Apply();
}
/// <summary>
/// Process the sample
/// </summary>
void Start()
{
// convert texture to cv image
Mat image = Unity.TextureToMat(this.sample);
// Detect faces
var gray = image.CvtColor(ColorConversionCodes.BGR2GRAY);
Cv2.EqualizeHist(gray, gray);
// detect matching regions (faces bounding)
Rect[] rawFaces = cascadeFaces.DetectMultiScale(gray, 1.1, 6);
// now per each detected face draw a marker and detect eyes inside the face rect
foreach (var faceRect in rawFaces)
{
var grayFace = new Mat(gray, faceRect);
if (requiredSize.Width > 0 && requiredSize.Height > 0)
{
grayFace = grayFace.Resize(requiredSize);
}
// now try to recognize the face:
// "confidence" here is actually a misguide. in fact, it's "distance from the sample to the closest known face" where
// exact metric is not disclosed in the docs, but checking returned values I found "confidence" to be like 70-100 for
// positive match with LBPH algo and more like 700-1200 for positive match with EigenFaces/FisherFaces. Unfortunately,
// all that data isn't much helpful for real life as you don't get adequate % of the confidence, the only thing you
// actually know is "less is better" with 0 being some "ideal match"
int label = -1;
double confidence = 0.0;
recognizer.Predict(grayFace, out label, out confidence);
bool found = confidence < 1200;
Scalar frameColor = found ? Scalar.LightGreen : Scalar.Red;
Cv2.Rectangle((InputOutputArray)image, faceRect, frameColor, 2);
int line = 0;
const int textPadding = 2;
const double textScale = 2.0;
string messge = String.Format("{0}", names[label], (int)confidence);
var textSize = Cv2.GetTextSize(messge, HersheyFonts.HersheyPlain, textScale, 1, out line);
var textBox = new Rect(
faceRect.X + (faceRect.Width - textSize.Width) / 2 - textPadding,
faceRect.Bottom,
textSize.Width + textPadding * 2,
textSize.Height + textPadding * 2
);
Cv2.Rectangle((InputOutputArray)image, textBox, frameColor, -1);
image.PutText(messge, textBox.TopLeft + new Point(textPadding, textPadding + textSize.Height), HersheyFonts.HersheyPlain, textScale, Scalar.Black, 2);
}
// Render texture
var texture = Unity.MatToTexture(image);
var rawImage = gameObject.GetComponent <UnityEngine.UI.RawImage>();
rawImage.texture = texture;
var transform = gameObject.GetComponent <UnityEngine.RectTransform>();
transform.sizeDelta = new UnityEngine.Vector2(image.Width, image.Height);
}
public static double?GetRotationAngle(string inputFileFolderPath, string inputFileName, bool showMessageBoxes, ImageBox iboxRaw, ImageBox iboxProcessed, out double msecElapsed)
{
double?rotationAngle = null;
msecElapsed = 0;
// Hough algo does a bad job detecting horizontal lines. So we rotate the image by a set amount before running the Hough.
double houghRotationOffsetAngle = 25.0;
try
{
Stopwatch stopWatch = new Stopwatch();
stopWatch.Start();
if (iboxProcessed != null)
{
iboxProcessed.Image = null;
iboxProcessed.Refresh();
}
Mat rotated = new Mat();
Mat src = new Mat(inputFileFolderPath + inputFileName, ImreadModes.Grayscale);
if (showMessageBoxes && iboxRaw != null)
{
iboxRaw.Image = OpenCvSharp.Extensions.BitmapConverter.ToBitmap(src);
}
// Not needed if we read as grayscale to start with.
//Mat src8UC1 = new Mat();
//src.ConvertTo(src8UC1, MatType.CV_8UC1);
// I'm not sure why we do the gauss - It seems like everyone does it, it's cheap, so we do it. ~Ed
Mat gauss = new Mat();
Cv2.GaussianBlur(src, gauss, new OpenCvSharp.Size(3, 3), 2, 2);
LogEvent("gauss", showMessageBoxes, gauss, iboxProcessed);
// An attempt to get the contrast across the image to be somewhat uniform.
CLAHE claheFilter = Cv2.CreateCLAHE(4, new OpenCvSharp.Size(10, 10));
Mat clahe = new Mat();
claheFilter.Apply(gauss, clahe);
LogEvent("clahe", showMessageBoxes, clahe, iboxProcessed);
// An attempt to get the contrast across the image to be somewhat uniform.
Mat hist = new Mat();
Cv2.EqualizeHist(gauss, hist);
LogEvent("hist", showMessageBoxes, hist, iboxProcessed);
// Grab a template from some middle part of the image. Eventually, the size and location of this
// template will be specified. It is very possible we'll have to grab multiple templates, as the
// location of the template may impact the accuracy of the rotation.
// e.g. - if the template is an image of a damaged device (which may happen at any location), the calculated
// rotation may be wrong. Testing is required.
// The locations where the template matches will create an image with lines that are offset from 0/90 degrees.
// This is because we can assume that the devices are orthogonal to one another, even if the image itself is
// offset rotationally.
Rect r1 = new Rect(new OpenCvSharp.Point(1000, 1000), new OpenCvSharp.Size(500, 300));
var roi = new Mat(clahe, r1);
Mat template = new Mat(new OpenCvSharp.Size(500, 300), MatType.CV_8UC1);
roi.CopyTo(template);
LogEvent("template", showMessageBoxes, template, iboxProcessed);
Mat templateMatch = new Mat();
Cv2.MatchTemplate(clahe, template, templateMatch, TemplateMatchModes.CCoeffNormed);
LogEvent("templatematch", showMessageBoxes, templateMatch, iboxProcessed);
Mat normalized = new Mat();
normalized = templateMatch.Normalize(0, 255, NormTypes.MinMax);
normalized.ConvertTo(normalized, MatType.CV_8UC1);
LogEvent("normalized template match", showMessageBoxes, normalized, iboxProcessed);
// This winnows down the number of matches.
Mat thresh = new Mat();
Cv2.Threshold(normalized, thresh, 200, 255, ThresholdTypes.Binary);
LogEvent("threshold template match", showMessageBoxes, thresh, iboxProcessed);
// rotate the image because hough doesn't work very well to find horizontal lines.
Mat rotatedThresh = new Mat();
Cv2E.RotateDegrees(thresh, rotatedThresh, houghRotationOffsetAngle);
LogEvent("rotatedThresh", showMessageBoxes, rotatedThresh, iboxProcessed);
Mat erode = new Mat();
Cv2.Erode(rotatedThresh, erode, new Mat());
LogEvent("erode", showMessageBoxes, erode, iboxProcessed);
LineSegmentPoint[] segHoughP = Cv2.HoughLinesP(rotatedThresh, 1, Math.PI / 1800, 2, 10, 600);
Mat imageOutP = new Mat(src.Size(), MatType.CV_8UC3);
// We're limiting the rotation correction to +/- 10 degrees. So we only care about hough lines that fall within 80 to 100 or 170 to 190
List <double> anglesNear90 = new List <double>();
List <double> anglesNear0 = new List <double>();
foreach (LineSegmentPoint s in segHoughP)
{
try
{
// Add lines to the image, if we're going to look at it.
if (showMessageBoxes)
{
imageOutP.Line(s.P1, s.P2, Scalar.White, 1, LineTypes.AntiAlias, 0);
}
var radian = Math.Atan2((s.P1.Y - s.P2.Y), (s.P1.X - s.P2.X));
var angle = ((radian * (180 / Math.PI) + 360) % 360);
// We rotated the image because the hough algo does a bad job with small horizontal lines. So we take that rotation back out here.
angle += houghRotationOffsetAngle;
angle -= 180;
if (angle > 80 && angle < 100)
{
anglesNear90.Add(angle);
if (showMessageBoxes)
{
imageOutP.Line(s.P1, s.P2, Scalar.Red, 1, LineTypes.AntiAlias, 0);
}
}
if (angle > -10 && angle < 10)
{
anglesNear0.Add(angle);
if (showMessageBoxes)
{
imageOutP.Line(s.P1, s.P2, Scalar.Orange, 1, LineTypes.AntiAlias, 0);
}
}
}
catch (Exception ex)
{
// there's always some infinity risk with atan, yes? Maybe. I don't want to fail on horizontal or vertical line edge cases.
}
}
double meanAngleNear0 = 0;
if (anglesNear0.Count > 0)
{
meanAngleNear0 = anglesNear0.Mean();
}
double meanAngleNear90 = 90;
if (anglesNear90.Count > 0)
{
meanAngleNear90 = anglesNear90.Mean();
}
// Use both the vertical and horizontal to calculate the image angle with a weighted average. It might be more accurate to use median instead of mean here.
rotationAngle = ((meanAngleNear0) * anglesNear0.Count + (meanAngleNear90 - 90) * anglesNear90.Count) / (anglesNear0.Count + anglesNear90.Count);
LogEvent("hough lines", showMessageBoxes, imageOutP, iboxProcessed);
stopWatch.Stop();
// Get the elapsed time as a TimeSpan value. Less than 400msec in debug mode via IDE.
TimeSpan ts = stopWatch.Elapsed;
msecElapsed = ts.TotalMilliseconds;
}
catch (Exception ex)
{
}
return(rotationAngle);
}
public static void DetectFacesOnImage(string sourceImagePath, string destImagePath)
{
var srcImage = new Mat(sourceImagePath);
// Cv2.ImShow("Source", srcImage);
Cv2.WaitKey(1); // do events
var grayImage = new Mat();
Cv2.CvtColor(srcImage, grayImage, ColorConversionCodes.BGRA2GRAY);
Cv2.EqualizeHist(grayImage, grayImage);
var cascade = new CascadeClassifier(@"./OpenCvSharp/haarcascade_frontalface_alt.xml");
var nestedCascade = new CascadeClassifier(@"./OpenCvSharp/haarcascade_eye_tree_eyeglasses.xml");
var faces = cascade.DetectMultiScale(
image: grayImage,
scaleFactor: 1.1,
minNeighbors: 2,
flags: HaarDetectionType.DoRoughSearch | HaarDetectionType.ScaleImage,
minSize: new Size(30, 30)
);
Console.WriteLine("Detected faces: {0}", faces.Length);
// Get accessories of the faces
var rnd = new Random();
var count = 1;
foreach (var faceRect in faces)
{
var detectedFaceImage = new Mat(srcImage, faceRect);
// Cv2.ImShow(string.Format("Face {0}", count), detectedFaceImage);
Cv2.WaitKey(1); // do events
// Draw rectangle arounf face
var color = Scalar.FromRgb(rnd.Next(0, 255), rnd.Next(0, 255), rnd.Next(0, 255));
Cv2.Rectangle(srcImage, faceRect, color, 2);
// Get gray image
var detectedFaceGrayImage = new Mat();
Cv2.CvtColor(detectedFaceImage, detectedFaceGrayImage, ColorConversionCodes.BGRA2GRAY);
var nestedObjects = nestedCascade.DetectMultiScale(
image: detectedFaceGrayImage,
scaleFactor: 1.1,
minNeighbors: 2,
flags: HaarDetectionType.DoRoughSearch | HaarDetectionType.ScaleImage,
minSize: new Size(30, 30)
);
Console.WriteLine("Nested Objects[{0}]: {1}", count, nestedObjects.Length);
// Draw circle around accessory
//foreach (var nestedObject in nestedObjects)
//{
// var center = new Point
// {
// X = (int)(Math.Round(nestedObject.X + nestedObject.Width * 0.5, MidpointRounding.ToEven) + faceRect.Left),
// Y = (int)(Math.Round(nestedObject.Y + nestedObject.Height * 0.5, MidpointRounding.ToEven) + faceRect.Top)
// };
// var radius = Math.Round((nestedObject.Width + nestedObject.Height) * 0.25, MidpointRounding.ToEven);
// Cv2.Circle(srcImage, center, (int)radius, color, thickness: 3);
//}
count++;
}
// Cv2.ImShow("Haar Detection", srcImage);
Cv2.WaitKey(1); // do events
// Save result image
srcImage.SaveImage(destImagePath);
Cv2.WaitKey(0);
Cv2.DestroyAllWindows();
srcImage.Dispose();
}
