public void NotSupportedUnicodeFileName()
{
Assert.True(File.Exists(LocalModelPath), $"'{LocalModelPath}' not found");
var unicodeFileName = Path.Combine(Path.GetDirectoryName(LocalModelPath) !, "🤣🍀.pb");
if (!File.Exists(unicodeFileName))
{
File.Copy(LocalModelPath, unicodeFileName, true);
}
// Check that ArgumentException(unicode unmappable char) does not occur.
if (RuntimeInformation.IsOSPlatform(OSPlatform.Windows))
{
var ex = Assert.Throws <OpenCVException>(() =>
{
using var net = CvDnn.ReadNet(unicodeFileName);
});
Assert.StartsWith("FAILED: fs.is_open(). Can't open", ex.Message, StringComparison.InvariantCulture);
Assert.Equal("cv::dnn::ReadProtoFromBinaryFile", ex.FuncName);
}
else
{
// No error
}
}
public void DetectAllText(string fileName)
{
const int InputWidth = 320;
const int InputHeight = 320;
const float ConfThreshold = 0.5f;
const float NmsThreshold = 0.4f;
// Load network.
using (Net net = CvDnn.ReadNet(Path.GetFullPath(LocalModelPath)))
using (Mat img = new Mat(fileName))
// Prepare input image
using (var blob = CvDnn.BlobFromImage(img, 1.0, new Size(InputWidth, InputHeight), new Scalar(123.68, 116.78, 103.94), true, false))
{
// Forward Pass
// Now that we have prepared the input, we will pass it through the network. There are two outputs of the network.
// One specifies the geometry of the Text-box and the other specifies the confidence score of the detected box.
// These are given by the layers :
// feature_fusion/concat_3
// feature_fusion/Conv_7/Sigmoid
var outputBlobNames = new string[] { "feature_fusion/Conv_7/Sigmoid", "feature_fusion/concat_3" };
var outputBlobs = outputBlobNames.Select(_ => new Mat()).ToArray();
net.SetInput(blob);
net.Forward(outputBlobs, outputBlobNames);
Mat scores = outputBlobs[0];
Mat geometry = outputBlobs[1];
// Decode predicted bounding boxes (decode the positions of the text boxes along with their orientation)
Decode(scores, geometry, ConfThreshold, out var boxes, out var confidences);
// Apply non-maximum suppression procedure for filtering out the false positives and get the final predictions
CvDnn.NMSBoxes(boxes, confidences, ConfThreshold, NmsThreshold, out var indices);
// Render detections.
Point2f ratio = new Point2f((float)img.Cols / InputWidth, (float)img.Rows / InputHeight);
for (var i = 0; i < indices.Length; ++i)
{
RotatedRect box = boxes[indices[i]];
Point2f[] vertices = box.Points();
for (int j = 0; j < 4; ++j)
{
vertices[j].X *= ratio.X;
vertices[j].Y *= ratio.Y;
}
for (int j = 0; j < 4; ++j)
{
Cv2.Line(img, (int)vertices[j].X, (int)vertices[j].Y, (int)vertices[(j + 1) % 4].X, (int)vertices[(j + 1) % 4].Y, new Scalar(0, 255, 0), 3);
}
}
ShowImagesWhenDebugMode(img);
}
}
public InferenceEngineDetector(string modelFile)
{
Net = CvDnn.ReadNet(Path.ChangeExtension(modelFile, ".bin"), Path.ChangeExtension(modelFile, ".xml"));
Net.SetPreferableBackend(Net.Backend.INFERENCE_ENGINE);
Net.SetPreferableTarget(Net.Target.OPENCL_FP16);
//Net.SetPreferableTarget(Net.Target.OPENCL_FP16);
//Net.SetPreferableBackend(Net.Backend.OPENCV);
OutputMat = new Mat();
}
public InferenceEngineDetector(string modelFile, Net.Backend backend, Net.Target target)
{
Net = CvDnn.ReadNet(Path.ChangeExtension(modelFile, ".bin"), Path.ChangeExtension(modelFile, ".xml"));
Net.SetPreferableTarget(target);
Net.SetPreferableBackend(backend);
Console.WriteLine($"Backend = '{backend}', Target = '{target}'");
//Net.SetPreferableBackend(Net.Backend.INFERENCE_ENGINE);
//Net.SetPreferableTarget(Net.Target.OPENCL_FP16);
//Net.SetPreferableTarget(Net.Target.OPENCL_FP16);
//Net.SetPreferableBackend(Net.Backend.OPENCV);
OutputMat = new Mat();
}
public Sbd(string modelPath, Size inputSize, int outputWidth, int outputHeight, string[] classNames)
{
net = CvDnn.ReadNet(modelPath);
if (net.Empty())
{
Console.WriteLine("fail to load model");
return;
}
net.SetPreferableBackend(Net.Backend.OPENCV);
net.SetPreferableTarget(Net.Target.CPU);
this.inputSize = inputSize;
this.outputWidth = outputWidth;
this.outputHeight = outputHeight;
this.classNames = classNames;
this.classCount = classNames.Length;
}
public void Run()
{
const string modelFace = "face-detection-adas-0001.bin";
const string modelFaceTxt = "face-detection-adas-0001.xml";
const string sampleImage = "sample.jpg";
const string outputLoc = "sample_output.jpg";
using var frame = Cv2.ImRead(sampleImage);
int frameHeight = frame.Rows;
int frameWidth = frame.Cols;
using var netFace = CvDnn.ReadNet(modelFace, modelFaceTxt);
netFace.SetPreferableBackend(Backend.INFERENCE_ENGINE);
netFace.SetPreferableTarget(Target.CPU);
using var blob = CvDnn.BlobFromImage(frame, 1.0, new OpenCvSharp.Size(672, 384), new OpenCvSharp.Scalar(0, 0, 0), false, false);
netFace.SetInput(blob);
using (var detection = netFace.Forward())
{
Mat detectionMat = new Mat(detection.Size(2), detection.Size(3), MatType.CV_32F, detection.Ptr(0));
for (int i = 0; i < detectionMat.Rows; i++)
{
float confidence = detectionMat.At <float>(i, 2);
if (confidence > 0.7)
{
int x1 = (int)(detectionMat.At <float>(i, 3) * frameWidth); //xmin
int y1 = (int)(detectionMat.At <float>(i, 4) * frameHeight); //ymin
int x2 = (int)(detectionMat.At <float>(i, 5) * frameWidth); //xmax
int y2 = (int)(detectionMat.At <float>(i, 6) * frameHeight); //ymax
OpenCvSharp.Rect roi = new OpenCvSharp.Rect(x1, y1, (x2 - x1), (y2 - y1));
roi = AdjustBoundingBox(roi);
Cv2.Rectangle(frame, roi, new Scalar(0, 255, 0), 2, LineTypes.Link4);
}
}
}
var finalOutput = outputLoc;
Cv2.ImWrite(finalOutput, frame);
}
public void Load()
{
Assert.True(File.Exists(LocalModelPath), $"'{LocalModelPath}' not found");
using var net = CvDnn.ReadNet(LocalModelPath);
}
public void Run()
{
const string modelFace = "face-detection-adas-0001.bin";
const string modelFaceTxt = "face-detection-adas-0001.xml";
const string modelHead = "head-pose-estimation-adas-0001.bin";
const string modelHeadTxt = "head-pose-estimation-adas-0001.xml";
const string sampleImage = "sample.jpg";
const string outputLoc = "sample_output.jpg";
var frame = Cv2.ImRead(sampleImage);
int frameHeight = frame.Rows;
int frameWidth = frame.Cols;
var netFace = CvDnn.ReadNet(modelFace, modelFaceTxt);
var netHead = CvDnn.ReadNet(modelHead, modelHeadTxt);
netFace.SetPreferableBackend(Net.Backend.INFERENCE_ENGINE);
netFace.SetPreferableTarget(Net.Target.CPU);
netHead.SetPreferableBackend(Net.Backend.INFERENCE_ENGINE);
netHead.SetPreferableTarget(Net.Target.CPU);
var blob = CvDnn.BlobFromImage(frame, 1.0, new OpenCvSharp.Size(672, 384), new OpenCvSharp.Scalar(0, 0, 0), false, false);
netFace.SetInput(blob);
using (var detection = netFace.Forward())
{
Mat detectionMat = new Mat(detection.Size(2), detection.Size(3), MatType.CV_32F, detection.Ptr(0));
for (int i = 0; i < detectionMat.Rows; i++)
{
float confidence = detectionMat.At <float>(i, 2);
if (confidence > 0.7)
{
int x1 = (int)(detectionMat.At <float>(i, 3) * frameWidth); //xmin
int y1 = (int)(detectionMat.At <float>(i, 4) * frameHeight); //ymin
int x2 = (int)(detectionMat.At <float>(i, 5) * frameWidth); //xmax
int y2 = (int)(detectionMat.At <float>(i, 6) * frameHeight); //ymax
OpenCvSharp.Rect roi = new OpenCvSharp.Rect(x1, y1, (x2 - x1), (y2 - y1));
roi = AdjustBoundingBox(roi);
Mat face = new Mat(frame, roi);
var blob2 = CvDnn.BlobFromImage(face, 1.0, new OpenCvSharp.Size(60, 60), new OpenCvSharp.Scalar(0, 0, 0), false, false);
netHead.SetInput(blob2);
IEnumerable <string> outNames = netHead.GetUnconnectedOutLayersNames();
IEnumerable <Mat> outputBlobs = new List <Mat>()
{
new Mat(), new Mat(), new Mat()
};
netHead.Forward(outputBlobs, outNames);
Point3f headAngles = HeadAngles(outputBlobs);
string printAngles = "Yaw " + headAngles.Y.ToString() + " | Pitch " + headAngles.X.ToString() + " | Roll " + headAngles.Z.ToString();
Cv2.Rectangle(frame, roi, new Scalar(0, 255, 0), 2, LineTypes.Link4);
Cv2.PutText(frame, printAngles, new OpenCvSharp.Point(10, 500), HersheyFonts.HersheyComplex, 1.0, new Scalar(0, 255, 0), 2);
}
}
}
var finalOutput = outputLoc;
Cv2.ImWrite(finalOutput, frame);
}
private void Form1_Load(object sender, EventArgs e)
{
var model = "opencv_face_detector_uint8.pb"; //tensorflow 학습된 모델 파일
var config = "opencv_face_detector.pbtxt"; // 구성 파일
var confidence = 0.0;
var x1 = 0.0;
var x2 = 0.0;
var y1 = 0.0;
var y2 = 0.0;
int cnt = 0;
video = new VideoCapture(1); //video 1번(ManyCam) 사용
if (!video.IsOpened()) //video 초기화 안될 시 error
{
Console.WriteLine("error");
return;
}
Net net = CvDnn.ReadNet(model, config);
if (net.Empty())
{
Console.WriteLine("error");
return;
}
frame = new Mat();
while (true)
{
video.Read(frame);
if (frame.Empty())
{
break;
}
//frame blob 입력 설정된 부분을 net
Mat blob = CvDnn.BlobFromImage(frame, 1.0, new OpenCvSharp.Size(300, 300), new Scalar(104, 177, 123));
net.SetInput(blob);
Mat res = net.Forward();
var detect = res.Reshape(1, res.Cols * res.Rows);
for (int i = 0; i < detect.Rows; i++) //Face recongition
{
confidence = detect.At <float>(i, 2);
if (confidence < 0.5)
{
break; //Face recongition 0.5 이하는 무시
}
//get center and width/height
x1 = Math.Round(detect.At <float>(i, 3) * frame.Cols); //좌표 계산
y1 = Math.Round(detect.At <float>(i, 4) * frame.Rows);
x2 = Math.Round(detect.At <float>(i, 5) * frame.Cols);
y2 = Math.Round(detect.At <float>(i, 6) * frame.Rows);
string Label = string.Format("Face:" + confidence); // 1. Face recongition을 시각화하는 작업
//2. confidence > 0.955 greenbox 생성
if (confidence > 0.955)
{
Cv2.Rectangle(frame, new OpenCvSharp.Point(x1, y1), new OpenCvSharp.Point(x2, y2), new Scalar(0, 255, 0));
Cv2.PutText(frame, Label, new OpenCvSharp.Point(x1, y1 - 1), OpenCvSharp.HersheyFonts.Italic, 0.8, new Scalar(0, 255, 0));
}
//2. 아닐 시 redbox 생성
else
{
Cv2.Rectangle(frame, new OpenCvSharp.Point(x1, y1), new OpenCvSharp.Point(x2, y2), new Scalar(0, 0, 255));
Cv2.PutText(frame, Label, new OpenCvSharp.Point(x1, y1 - 1), OpenCvSharp.HersheyFonts.Italic, 0.8, new Scalar(0, 0, 255));
}
}
Cv2.ImShow("FACE ID", frame);
if (confidence > 0.997910)
{
cnt++;
if (cnt >= 30) //30번 인식시 파일 save
{
_saveName = "C:/images/" + DateTime.Now.ToString("yyyy/MM/dd_hh_mm_ss") + ".jpeg";
Cv2.ImWrite(_saveName, frame);
video.Release();
Cv2.DestroyAllWindows();
break;
}
}
if (Cv2.WaitKey(1) == 27) //아스키코드 27 : esc
{
break;
}
}
Delay(100);
inCvImage = Cv2.ImRead(_saveName); //2번
Cv2.Transpose(inCvImage, inCvImage);
outCvImage = new Mat();
Cv2.CvtColor(inCvImage, outCvImage, ColorConversionCodes.BGR2GRAY);
Cv2.EqualizeHist(outCvImage, outCvImage);
Cv2ToOutImage();
//사각박스 내부만 나오게 설정
for (int rgb = 0; rgb < 3; rgb++)
{
for (int i = 0; i < _outH; i++)
{
for (int k = 0; k < _outW; k++)
{
if ((x1 <= i && i <= x2) && (y1 <= k && k <= y2))
{
_outImage[rgb, i, k] = (byte)(255 - _outImage[rgb, i, k]);
}
else
{
_outImage[rgb, i, k] = 0;
}
}
}
}
DisplayImage();
connStr = "Server=127.0.0.1; Uid=root; Pwd=1234;Database=HYDB;CHARSET=UTF8";
conn = new MySqlConnection(connStr);
//db로 전달
dbUpload();
CCTV ma = new CCTV();
ma.ShowDialog();
Close();
} //1. 얼굴인식 후 저장 및 blob 업로드