Ask Your Question

Marwen Z's profile - activity

2018-07-24 02:31:57 -0500 received badge  Popular Question (source)
2017-09-21 16:36:03 -0500 received badge  Notable Question (source)
2016-06-08 08:27:44 -0500 received badge  Famous Question (source)
2016-04-25 22:44:12 -0500 received badge  Notable Question (source)
2015-12-30 03:58:46 -0500 received badge  Popular Question (source)
2014-12-09 13:51:47 -0500 marked best answer object recognition in real time on Android with FAST detector

The aim of this work is to extract real-time key points from an existing image in a video scene processFrame has the execution that does not work, due to match, is that I can just display the key points of correspondence in the form of circles on the image in real time

   class Sample1View extends SampleViewBase {
    public static final int     VIEW_MODE_RGBA   = 0;
    public static final int  VIEW_MODE_BLUE  = 1; 
    public static final int  VIEW_MODE_YELLOW = 2;
    public static final int  VIEW_MODE_DE = 3;
    private Mat mYuv;
    private Mat mRgba;
    private Mat mGraySubmat;
    private Mat mResult;
    private Mat mIntermediateMat;
 private Bitmap mBitmap;
 private int mViewMode;
 private Mat mColor;
 private Mat mHsv;
 TimingLogger timings;
   private Mat img1;
    private Mat descriptors;
    private MatOfKeyPoint keypoints;
    private FeatureDetector detector;
    private DescriptorExtractor descriptor;
    private DescriptorMatcher matcher;

 private static final String TAG ="Sample::View";

    public Sample1View(Context context) {
        mViewMode = VIEW_MODE_RGBA;

 try {
 img1=Utils.loadResource(getContext(), R.drawable.wings);
 } catch (IOException e) {
 // TODO Auto-generated catch block
 Log.w("Activity::LoadResource","Unable to load resource R.drawable.wings");
        descriptors = new Mat();
        keypoints = new MatOfKeyPoint();
        detector = FeatureDetector.create(FeatureDetector.FAST);
        detector.detect(img1, keypoints);
        descriptor = DescriptorExtractor.create(DescriptorExtractor.ORB);
        descriptor.compute(img1, keypoints, descriptors);
        matcher = DescriptorMatcher.create(DescriptorMatcher.BRUTEFORCE_HAMMING);

 protected void onPreviewStarted(int previewWidth, int previewHeight) {
     Log.i(TAG, "preview Started");
     synchronized (this) {
         mYuv = new Mat(getFrameHeight() + getFrameHeight() / 2, getFrameWidth(), CvType.CV_8UC1);
         mGraySubmat = mYuv.submat(0, getFrameHeight(), 0, getFrameWidth());
         mRgba = new Mat();
         mIntermediateMat = new Mat();
         mBitmap = Bitmap.createBitmap(previewWidth, previewHeight, Bitmap.Config.ARGB_8888); 
         mHsv = new Mat();
         mColor = new Mat();
         mResult = new Mat();

 protected void onPreviewStopped() {
     Log.i(TAG, "preview Stopped");
     if(mBitmap != null) {

  synchronized (this) {
            // Explicitly deallocate Mats
            if (mYuv != null)
            if (mRgba != null)
            if (mGraySubmat != null)
            if (mIntermediateMat != null)
            mYuv = null;
            mRgba = null;
            mGraySubmat = null;
            mIntermediateMat = null;
            if (mResult != null)
            if (mHsv != null)
            if (mColor != null)
            mColor = null;
         mResult = null;
         mHsv = null;
    protected Bitmap processFrame(byte[] data) {

        mYuv.put(0, 0, data);
        final int viewMode = mViewMode;


      MatOfKeyPoint mKeyPoints = new MatOfKeyPoint();
    MatOfDMatch  matches = new MatOfDMatch();
      detector.detect(mGraySubmat, mKeyPoints);
      descriptor.compute(mGraySubmat, mKeyPoints, mIntermediateMat);
      matcher.match(mIntermediateMat, descriptors, matches);
      mIntermediateMat2.create(resultSize, CvType.CV_8UC1);

      Features2d.drawMatches(mGraySubmat, mKeyPoints, mGraySubmat, mKeyPoints, matches, mIntermediateMat2);
      Imgproc.resize(mIntermediateMat2, mIntermediateMat2, mRgba.size());
      Imgproc.cvtColor(mIntermediateMat2, mRgba, Imgproc.COLOR_RGBA2BGRA, 4);
    Bitmap bmp = mBitmap;

    try {
      Utils.matToBitmap(mRgba, bmp);

    } catch(Exception e) {
        Log.e("org.opencv.samples.*", "Utils.matToBitmap() throws an exception: " + e.getMessage());
        bmp = null;
    return bmp;

    public void setViewMode(int viewMode) {
     mViewMode = viewMode;
2014-11-04 01:08:19 -0500 received badge  Notable Question (source)
2014-09-26 15:58:19 -0500 received badge  Popular Question (source)
2014-03-08 20:03:28 -0500 received badge  Popular Question (source)
2013-05-14 11:18:17 -0500 commented question How to perform Pose estimation for 3D object rendering in Android

did you found any solution

2013-05-06 03:13:43 -0500 commented question POSE estimation in OpenCv Java using cvFindExtrinsicsCameraParams2

Thank you, now I'm in the stage of registration of 3d graphics objects on the recognized object, that's why I did it like this

1-calibration of the camera

2-2D-3D matches

3-Calculating the pose

4-Registration of virtual object

i'm in step 2 can you give me ideas

2013-05-04 08:36:12 -0500 commented question POSE estimation in OpenCv Java using cvFindExtrinsicsCameraParams2

me too am in the process of implementing this algorithm and I'm referring to this book mastering_opencv_with_practical_computer_vision_projects , but until now I can not solve it

2013-04-30 06:32:58 -0500 answered a question superimpose 3D graphics objects on the recognized object

I do like this

1-calibration of the camera
2-2D-3D matches
3-Calculating the pose
4-Registration of virtual object

is that possible ? how to do it simply ? otherwise any help.

2013-04-29 15:35:56 -0500 commented answer object recognition in real time on Android with FAST detector
2013-04-29 08:58:25 -0500 commented answer object recognition in real time on Android with FAST detector

@StevenPuttemans after the recognition phase I want to insert a 3D object on the object detected, so what method to do this ? i don't wont to use OpenGL ES

2013-04-29 05:36:02 -0500 commented answer detect all objects in real-time with Android

@Mathieu Barnachon how can i superimpose object 3d in recognized object please guide me

2013-04-28 10:56:35 -0500 asked a question superimpose 3D graphics objects on the recognized object

after the phase of object recognition I took the best point from matching and would like a registration of 3D objects Please guide me on the big lines PS: I have read articles on the pose of the camera and camera calibration

2013-04-22 04:03:26 -0500 asked a question minVal & maxVal = 0.0 with TemplateMatching Android

Hi i use this code link text but it gives me minVal=maxVal=0.0 !

public void TemplateMatching(Mat mFind,Bitmap dst)

        Mat  Input = new Mat(mFind.width(), mFind.height(), CvType.CV_8UC4);
        Utils.bitmapToMat(dst, Input);
        Mat  mResult8u = new Mat(mFind.width(), mFind.height(), CvType.CV_8UC4);
        Mat mResult9u = new Mat(mFind.width(), mFind.height(), CvType.CV_8UC4);
        Mat mResult = new Mat(mFind.width(), mFind.height(), CvType.CV_8UC4);

        Imgproc.matchTemplate(mFind, Input, mResult, Imgproc.TM_CCOEFF_NORMED);
        Core.normalize(mResult, mResult9u, 0, 255, Core.NORM_MINMAX, CvType.CV_8U); 
        MinMaxLocResult locRes = Core.minMaxLoc(mResult9u);  
        double minVal = locRes.minVal;  
        Point minLoc = locRes.minLoc;
        double maxVal = locRes.maxVal;  
        Point maxLoc = locRes.maxLoc;

        Point point = new Point();

             double thresholds=0.08;
Log.w("max val", maxVal+"");
Log.w("min val", minVal+"");
        Core.rectangle(mFind, maxLoc,point, new Scalar(0, 255, 0, 255), 3);

        Bitmap bmp3= Bitmap.createBitmap(mFind.cols(), mFind.rows(),Bitmap.Config.ARGB_8888);
        Utils.matToBitmap(mFind, bmp3);     


2013-04-22 03:33:19 -0500 commented question Template Matching with Multiple Occurance

Hi i use the same code but it gives me maxVal =minVal =0.0 !!

2013-04-20 15:37:43 -0500 received badge  Self-Learner (source)
2013-04-20 14:13:01 -0500 commented question How to return a Rect object from native code to Java

what you're looking about is exist in the sample face detection in FdActivity

2013-04-20 14:02:35 -0500 answered a question detect object from images

@Guanta, ok I'll try it, what you suppose like histSize ? about recognize object have you any idea how to find DIST_LIMIT factor, I want to extract exactly the same picture, but I can not, or you have another solution (efficace condition) to find the base image.

my solution is to compare the descriptors of each image in the database and return the image to the greater length of good matches but it is not efficace !! here the code :

    matcher.match(descriptor2,descriptor1, matches);
    int DIST_LIMIT = 80;
    List<DMatch> matchesList = matches.toList();
    List<DMatch> matches_final= new ArrayList<DMatch>();
    for(int i=0; i<matchesList.size(); i++)
       if(matchesList .get(i).distance <= DIST_LIMIT){

     Log.w("good matches", matches_final.size()+"");
     return matches_final.size();
2013-04-16 04:50:23 -0500 answered a question detect all objects in real-time with Android

@Mathieu Barnachon, Hi after detecting an object from video using a spefic color, i want to recognize this object from images, so i use compare histogram but it gives me double d = 1.0 for all images ! then i try to use detect feature (ORB,FAST,SIFT) then i count the good matching and retrieve the image who had the biggest length of good_matching but it return me the false image. i don't exactly how to retrieve the approach image,

2013-04-16 04:46:09 -0500 commented answer image comparison

Hi after detecting an object from video using a spefic color, i want to recognize this object from images, so i use compare histogram but it gives me double d = 1.0 for all images ? then i try to use detect feature (ORB,FAST,SIFT) then i count the good matching and retrieve the image who had the biggest length of good_matching but it return me the false image.

2013-04-16 04:41:18 -0500 commented answer Which matcher is best for SURF?

@Szippy , do you solve your problem ?

2013-04-16 04:27:33 -0500 asked a question detect object from images

Hi after detecting an object from video using a spefic color, i want to recognize this object from images, so i use compare histogram but it gives me double d = 1.0 for all images ? then i try to use detect feature (ORB,FAST,SIFT) then i count the good matching and retrieve the image who had the biggest length of good_matching but it return me the false image.

for compare hisogram i do like this : i travel all elements image : images

List<Mat> imagesList=new ArrayList<Mat>();
            List<Mat> imagesList1=new ArrayList<Mat>();

            MatOfInt channels=new MatOfInt(0);
            Mat hist=new Mat();
            Mat hist1=new Mat();
            MatOfInt histSize=new MatOfInt(50);

            float hrangesArray[]={0.0f,255.0f};
            MatOfFloat ranges=new MatOfFloat(hrangesArray);

            Mat mask=new Mat();
            Imgproc.calcHist(imagesList, channels,mask, hist, histSize, ranges);
            Imgproc.calcHist(imagesList1, channels,mask, hist1, histSize, ranges);

          double i=  Imgproc.compareHist(hist1, hist, Imgproc.CV_COMP_BHATTACHARYYA);

and for the detecting feature i proced like in the tutorial.

but i don't know where is the problem ? thanks for help me

2013-04-09 12:17:50 -0500 commented answer Convert Mat to MatOfByte in Android !=0 mat.toString gives me detail of the mat

2013-04-09 10:55:26 -0500 commented answer Convert Mat to MatOfByte in Android

it gives me an empty Byte data

2013-04-08 04:07:10 -0500 commented answer Convert Mat to MatOfByte in Android

yes i try it but not working, i found another solution is to convert Mat to Bitmap then to Byte[]

Utils.matTobitmap(mRgba,bitmap); ByteArrayOutputStream os = new ByteArrayOutputStream(); bitmap.compress(CompressFormat.PNG, 100, os); bytes[] data= os.toByteArray(); //the opposite case : BitmapFactory.decodeByteArray(bytes, 0, bytes.length);

    How you thnig about it ? it's efficace  !
2013-04-07 12:36:02 -0500 commented answer convert Mat CV_8UC4 to byte[] Android

from a mRgba i want to convert it to a byte array because i want to store it in DB (ormlite) i follow this but don't work yes but why ? i didn't inderstand what you mean , i want just to convert a Mat (CV_8UC4) to byte array

2013-04-07 10:47:57 -0500 asked a question convert Mat CV_8UC4 to byte[] Android

Hi i try to convert a Mat to Byte then the inverse so here is the code

 frame = new Mat(height, width, CvType.CV_8UC4);
        // traitement ...
         byte[] data = = new byte[(int) ( *  frame.channels())];
            frame.get(0, 0, data); 
        // but data.length=0
2013-04-07 06:45:43 -0500 commented answer Convert Mat to MatOfByte in Android

@Andrey Pavlenko i want to convert a Mat CV_8UC4 to byte[] but don't work any of those solution!

2013-03-31 05:05:22 -0500 commented answer Is there any programs to recognize objects?

@StevenPuttemans thank you for the explanation, I have a small question to ask: from a video capture and a set of images (sqlite) I want to set the image that is closest to it I detect matches the two images and then using a technique such as Histogram to define closest distance

2013-03-31 03:47:00 -0500 commented question *java* api Histogram calculation

Hi have you find a solution !!

2013-03-25 11:25:24 -0500 asked a question select square from position using pointPolygonTest in Android

The purpose of this method is when I click a square with OnTouch (x,y) in a video he displayed it in another view Here are the steps:

initially an image and a set of square (time varying)

if the user clicks in a position then the algorithm check if he clicked on a square or not

if yes it displays the square in another view using a mask (ROI) else display the initial image the problem that working in console but in Android with jni don't response,

my question is how can i optimize this algorithm or found another way

it can be a performance problem => response time

void displaySquares(  Mat& image, vector<vector<Point> >& squares,int xa,int ya )

bool trouv=false;
vector<Rect> boundRect(squares.size()-squares.size()+1);
for(size_t i = 0; i < squares.size()/3; i++ )
vector<Point2f> points;
  for(size_t j = 0; j < 4; j++ )


if(pointPolygonTest((Mat)points, Point2f(xa,ya), false)>=0)

boundRect[0] = boundingRect( Mat(squares[i]) );

Mat roi(image,boundRect[0]);
Rect boundrect=boundRect[0];
rectangle(image, cv::Point(boundrect.x, boundrect.y),
         cv::Point(boundrect.x + boundrect.width, boundrect.y + boundrect.height), Scalar(0,255,0));

2013-03-23 09:23:09 -0500 received badge  Teacher (source)
2013-03-23 03:17:13 -0500 answered a question native opencv for android project

no , is your reference to ? LOCAL_PATH := $(call my-dir)

include $(CLEAR_VARS)

include ../../sdk/native/jni/

LOCAL_MODULE := Xxx LOCAL_SRC_FILES := Xxxx.cpp LOCAL_LDLIBS += -llog -ldl


and do you add

${NDKROOT}/platforms/android-9/arch-arm/usr/include ${NDKROOT}/sources/cxx-stl/gnu-libstdc++/4.6/include ${NDKROOT}/sources/cxx-stl/gnu-libstdc++/4.6/libs/armeabi-v7a/include ${ProjDirPath}/../../sdk/native/jni/include

( Open Project Properties -> C/C++ General -> Paths and Symbols and add the following Include paths for C++: )

2013-03-22 06:10:32 -0500 asked a question recognizing objects c++ Android

Hi , i just copied the code Features2D + Homography to find a known object from
on my Android project 2 inupts : a known object(Bitmap from drawable and i convert it to Mat) and an input frame but in Logcat it show me "Preview Frame received. Need to create MAT and deliver it to clients" BLACK FRAME

#include <jni.h>
#include <opencv2/core/core.hpp>
#include <opencv2/objdetect/objdetect.hpp>
#include <opencv2/imgproc/imgproc.hpp>
#include <opencv2/features2d/features2d.hpp>
#include <opencv2/nonfree/nonfree.hpp>
#include <opencv2/highgui/highgui.hpp>
#include <vector>
#include <android/log.h>

#define LOG_TAG "native"
#define LOGI(...) __android_log_print(ANDROID_LOG_INFO,LOG_TAG,__VA_ARGS__)
#define LOGE(...) __android_log_print(ANDROID_LOG_ERROR,LOG_TAG,__VA_ARGS__)
using namespace cv;
cv::Mat tmp, nomask, mask;

extern "C" {
JNIEXPORT void JNICALL Java_com_example_features_MainActivity_setup(JNIEnv*, jobject,
                                                                    jint rows, jint cols) {
  tmp.create(rows, cols, CV_8UC1);
  nomask = cv::Mat();
  mask = cv::Mat::zeros(cv::Size(cols, rows), CV_8UC1); // swap rows <-> cols
  cv::Mat roi(mask, cv::Rect(cols/2 - cols/4, rows/2 - rows/4, cols/2, rows/2));
  LOGI("rect: (%d %d)", roi.rows, roi.cols);
  roi = cv::Scalar(255);

JNIEXPORT void JNICALL Java_com_example_features_MainActivity_FindFeatures(JNIEnv*, jobject, jlong addrGray, jlong addrRgba)

        Mat& img_object  = *(Mat*)addrGray;
        Mat& img_scene = *(Mat*)addrRgba;
        int minHessian = 400;

          OrbFeatureDetector detector( minHessian );

        vector<KeyPoint> keypoints_object, keypoints_scene;

          detector.detect( img_object, keypoints_object );
          detector.detect( img_scene, keypoints_scene );

          //-- Step 2: Calculate descriptors (feature vectors)
          OrbDescriptorExtractor extractor;

          Mat descriptors_object, descriptors_scene;

          extractor.compute( img_object, keypoints_object, descriptors_object );
          extractor.compute( img_scene, keypoints_scene, descriptors_scene );

          //-- Step 3: Matching descriptor vectors using FLANN matcher
          FlannBasedMatcher   matcher;
          vector< DMatch > matches;
       matcher.match( descriptors_object, descriptors_scene, matches );

       //  drawKeypoints(img_scene, keypoints_scene,img_scene);
        double max_dist = 0; double min_dist = 100;

          //-- Quick calculation of max and min distances between keypoints
          for( int i = 0; i < descriptors_object.rows; i++ )
          { double dist = matches[i].distance;
            if( dist < min_dist ) min_dist = dist;
            if( dist > max_dist ) max_dist = dist;

          //-- Draw only "good" matches (i.e. whose distance is less than 3*min_dist )
          vector< DMatch > good_matches;

          for( int i = 0; i < descriptors_object.rows; i++ )
          { if( matches[i].distance < 3*min_dist )
             { good_matches.push_back( matches[i]); }

          Mat img_matches;
          drawMatches( img_object, keypoints_object, img_scene, keypoints_scene,
                       good_matches, img_matches, Scalar::all(-1), Scalar::all(-1),
                       vector<char>(), DrawMatchesFlags::NOT_DRAW_SINGLE_POINTS );

          //-- Localize the object
          vector<Point2f> obj;
          vector<Point2f> scene;

          for( int i = 0; i < good_matches.size(); i++ )
            //-- Get the keypoints from the good matches
            obj.push_back( keypoints_object[ good_matches[i].queryIdx ].pt );
            scene.push_back( keypoints_scene[ good_matches[i].trainIdx ].pt );

     //   Mat H = findHomography( obj, scene, CV_RANSAC );

          //-- Get the corners from the image_1 ( the object to be "detected" )
         vector<Point2f> obj_corners(4);
          obj_corners[0] = cvPoint(0,0); obj_corners[1] = cvPoint( img_object.cols, 0 );
          obj_corners[2] = cvPoint( img_object.cols, img_object.rows ); obj_corners[3] = cvPoint( 0, img_object.rows );
          vector<Point2f> scene_corners(4);

     //     perspectiveTransform( obj_corners, scene_corners, H);

          //-- Draw lines between the corners (the mapped object in the scene - image_2 )
          line( img_matches, scene_corners[0] + Point2f( img_object.cols, 0), scene_corners[1] + Point2f( img_object.cols, 0), Scalar(0, 255, 0), 4 );
          line( img_matches, scene_corners[1] + Point2f( img_object.cols, 0), scene_corners[2] + Point2f( img_object.cols, 0 ...
2013-03-21 09:03:52 -0500 answered a question locate the object recognition Android OpenCV

i solve the problem it was simple :p but same thing that give me things that not exist in fisrt image

      List<DMatch> goodMatchesList = new ArrayList<DMatch>();

    DMatch dm12[]=matches12.toArray();
    DMatch dm21[]=matches21.toArray();
    for(int i=0;i<dm12.length;i++)
     DMatch forward = dm12[i];  
        DMatch backward = dm21[forward.trainIdx];
        if( backward.trainIdx == forward.queryIdx ) 
            goodMatchesList.add(forward) ;


    matcher.match(descriptors,mIntermediateMat, matches);

    mIntermediateMat2.create(resultSize, CvType.CV_8UC1);

    Features2d.drawMatches( mGraySubmat, mKeyPoints, img1, keypoints,matches, mIntermediateMat2,GREEN, RED,  MATCH_MASK, Features2d.NOT_DRAW_SINGLE_POINTS);
2013-03-21 08:34:40 -0500 commented question locate the object recognition Android OpenCV

@StevenPuttemans my problem is i drawFeatures for a good matching so i need to create a new MatOfDmatch to push good matching but i didn't find a good code for that because Feature2d.drawMatches() need to MatOfDmatch and not List<Dmatch> did you have a solution that create a MatOfDmatch from others MatOfDmatch

2013-03-21 06:39:04 -0500 edited answer locate the object recognition Android OpenCV

@Mathieu Barnachon i procced like this

 MatOfDMatch matches, matches12, matches21;
    matcher.match( descriptors1, descriptors2, matches12 );
    matcher.match( descriptors2, descriptors1, matches21 );
 // iterate matches12
        DMatch forward = matches12[i];  // error convert from c++ to java
        DMatch backward = matches21[forward.trainIdx]; 
        if( backward.trainIdx == forward.queryIdx ) 
         matches.push_back( forward );  // error convert from c++ to java
    get the good matches  then draw it
     Features2d.drawMatches( mGraySubmat, mKeyPoints,img1, keypoints, matches, mIntermediateMat2,GREEN, RED,  MATCH_MASK, Features2d.NOT_DRAW_SINGLE_POINTS);