Skip to content

Instantly share code, notes, and snippets.

@vumbumy

vumbumy/cpp_face_detect_converted.java

Last active July 6, 2016 06:30
Show Gist options

  • Select an option

    Learn more about clone URLs
  • Save vumbumy/e8fadc582c4ab99df0f8b214959c9a70 to your computer and use it in GitHub Desktop.

Select an option

Learn more about clone URLs
Save vumbumy/e8fadc582c4ab99df0f8b214959c9a70 to your computer and use it in GitHub Desktop.
Download ZIP
Raw
cpp_face_detect_converted.java
public class MainActivity extends Activity implements CvCameraViewListener2 {
private CameraBridgeViewBase mOpenCvCameraView;
private static CascadeClassifier face_cascade; //계산을 위한 변수(얼굴용)
private static CascadeClassifier eyes_cascade; //계산을 위한 변수(눈용)
MatOfRect faces;
Mat mRgba; //컬러처리를 위한 화면 기본 변수
private BaseLoaderCallback mLoaderCallback = new BaseLoaderCallback(this) {
@Override
public void onManagerConnected(int status) {
switch (status) {
case LoaderCallbackInterface.SUCCESS:
{
face_cascade = loadCascade(
R.raw.haarcascade_frontalface_alt2,
"haarcascade_frontalface_alt2.xml"
);
//OpenCV가 제공하는 학습된 얼굴를 데이터 불러옴.
eyes_cascade = loadCascade(
R.raw.haarcascade_eye_tree_eyeglasses,
"haarcascade_eye_tree_eyeglasses.xml"
);
//OpenCV가 제공하는 학습된 눈의 데이터 불러옴.
mOpenCvCameraView.enableView();
} break;
default:
{
super.onManagerConnected(status);
} break;
}
}
};
private CascadeClassifier loadCascade(int RID, String fileName)
{
//파일 로드를 위한 함수
CascadeClassifier mJavaDetector = null;
try {
// load cascade file from application resources
InputStream is = getResources().openRawResource(RID);
File cascadeDir = getDir("cascade", Context.MODE_PRIVATE);
File mCascadeFile = new File(cascadeDir, fileName);
FileOutputStream os = new FileOutputStream(mCascadeFile);
byte[] buffer = new byte[4096];
int bytesRead;
while ((bytesRead = is.read(buffer)) != -1) {
os.write(buffer, 0, bytesRead);
}
is.close();
os.close();
mJavaDetector = new CascadeClassifier(mCascadeFile.getAbsolutePath());
if (mJavaDetector.empty()) {
Log.e("", "Failed to load cascade classifier");
mJavaDetector = null;
} else
Log.i("",
"Loaded cascade classifier from " + mCascadeFile.getAbsolutePath());
cascadeDir.delete();
} catch (IOException e) {
e.printStackTrace();
Log.e("", "Failed to load cascade. Exception thrown: " + e);
}
return mJavaDetector;
}
@Override
protected void onCreate(Bundle savedInstanceState) {
super.onCreate(savedInstanceState);
getWindow().addFlags(WindowManager.LayoutParams.FLAG_KEEP_SCREEN_ON);
setContentView(R.layout.activity_main);
mOpenCvCameraView =
(CameraBridgeViewBase)findViewById(
R.id.tutorial1_activity_java_surface_view
);
mOpenCvCameraView.setVisibility(SurfaceView.VISIBLE);
mOpenCvCameraView.setCvCameraViewListener(this);
}
public void onResume() {
super.onResume();
OpenCVLoader.initAsync(OpenCVLoader.OPENCV_VERSION_2_4_3, this, mLoaderCallback);
};
@Override
public boolean onCreateOptionsMenu(Menu menu) {
// Inflate the menu; this adds items to the action bar if it is present.
getMenuInflater().inflate(R.menu.main, menu);
return true;
}
@Override
public boolean onOptionsItemSelected(MenuItem item) {
// Handle action bar item clicks here. The action bar will
// automatically handle clicks on the Home/Up button, so long
// as you specify a parent activity in AndroidManifest.xml.
int id = item.getItemId();
if (id == R.id.action_settings) {
return true;
}
return super.onOptionsItemSelected(item);
}
@Override
public void onCameraViewStarted(int width, int height) {
mRgba = new Mat();
faces = new MatOfRect();
}
@Override
public void onCameraViewStopped() {
// TODO Auto-generated method stub
}
@Override
public Mat onCameraFrame(CvCameraViewFrame inputFrame) {
Core.flip(inputFrame.rgba(), mRgba, 1);
mRgba = detectAndDraw(mRgba,face_cascade,eyes_cascade,1,true);
return mRgba;
}
public Mat detectAndDraw(
Mat src,
CascadeClassifier cascade,
CascadeClassifier nestedCascade,
double scale, boolean tryflip) {
//cpp sample source
int i = 0;
double t = 0;
//vector<Rect> faces, faces2;
List<Rect> faces = new ArrayList<Rect>();
MatOfRect faces2 = new MatOfRect();
/*
* const static Scalar colors[] = { CV_RGB(0,0,255),
* CV_RGB(0,128,255),
* CV_RGB(0,255,255),
* CV_RGB(0,255,0),
* CV_RGB(255,128,0),
* CV_RGB(255,255,0),
* CV_RGB(255,0,0),
* CV_RGB(255,0,255)};
*/
ArrayList<Scalar> colors = new ArrayList<Scalar>();
colors.add(new Scalar(00,0,255));
colors.add(new Scalar(0,128,255));
colors.add(new Scalar(0,255,255));
colors.add(new Scalar(0,255,0));
colors.add(new Scalar(255,128,0));
colors.add(new Scalar(255,255,0));
colors.add(new Scalar(255,0,0));
colors.add(new Scalar(255,0,255));
Mat gray = new Mat();
//smallImg( cvRound (img.rows/scale), cvRound(img.cols/scale), CV_8UC1 );
Mat smallImg = new Mat(
(int)Math.round(src.rows()/scale),
(int)Math.round(src.cols()/scale), CvType.CV_8UC1
);
Imgproc.cvtColor(src, gray, Imgproc.COLOR_BGRA2GRAY);
Imgproc.resize(gray, smallImg, smallImg.size(),0,0,Imgproc.INTER_LINEAR);
Imgproc.equalizeHist(smallImg, smallImg);
MatOfRect mr = new MatOfRect();
mr.fromList(faces);
cascade.detectMultiScale(smallImg, mr,
1.1, 2, 0,new Size(30, 30), new Size());
if(tryflip){
Core.flip(smallImg, smallImg, 1);
cascade.detectMultiScale(smallImg, faces2,
1.1, 2, 0,new Size(30, 30), new Size());
Rect[] lr = faces2.toArray();
for(Rect r : lr){
faces.add(new Rect(
smallImg.cols()-r.x-r.width,
r.y,r.width,r.height)
);
}
}
for(Rect r : faces)
{
i++;
MatOfRect nestedObjects = new MatOfRect();
Point center = new Point();
Scalar color = colors.get(i%8);
int radius;
double aspect_ratio = (double)r.width/r.height;
if( 0.75 < aspect_ratio && aspect_ratio < 1.3 )
{
//cvRound -> (int)Math.round
center.x = (int)Math.round((r.x + r.width*0.5)*scale);
center.y = (int)Math.round(Math.round((r.y + r.height*0.5)*scale));
radius = (int)Math.round((r.width + r.height)*0.25*scale);
Core.circle( src, center, radius, color, 3, 8, 0 );
}
else
Core.rectangle(src,
new Point(
(int)Math.round(r.x*scale),
(int)Math.round(r.y*scale)
),
new Point(
(int)Math.round((r.x + r.width-1)*scale),
(int)Math.round((r.y + r.height-1)*scale)
),
color, 3, 8, 0);
if( nestedCascade.empty() )
continue;
//smallImgROI = smallImg(*r);
Mat smallImgROI = new Mat(src,r);
nestedCascade.detectMultiScale(
smallImgROI,
nestedObjects,
1.1, 2, 0,
new Size(30, 30), new Size());
for(Rect nr : nestedObjects.toList())
{
center.x = (int)Math.round((r.x + nr.x + nr.width*0.5)*scale);
center.y = (int)Math.round((r.y + nr.y + nr.height*0.5)*scale);
radius = (int)Math.round((nr.width + nr.height)*0.25*scale);
Core.circle( src, center, radius, color, 3, 8, 0 );
}
}
return src;
}
}
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment