kzbr93's profile - activity

@yes123, could you please explain more

@berak ok lets make it simple, i store the image of the object and try to track it, in a video !! i am trying to do it b

eventually its making that green color bounding box in whole of my current frame, then whats the use of those feature ma

@berak sorry i think i am not able to put my question properly, its not just about human faces, and i know that SURF wor

@berak sorry i think i am not able to put my question properly, its not just about human faces, and i know that SURF wor

how to make bounding box on surf features ? Hello all, i am working on SURF features and i will be glad if someone coul

i have video frames, is there then any other way i can make a bounding box on objects approaching near ??

@berak thanks for the reply, but i was wondering what is the use of feature points if we cannot distinguish those with r

how to make bounding box on surf features ? Hello all, i am working on SURF features and i will be glad if someone coul

how to make bounding box on surf features ? Hello all, i am working on SURF features and i will be glad if someone coul

it should not something impossible, they came out with algorithm which is tested... so i think its just a matter of time..

what i could do is, compare the previous frame with that of current, and do a template matching, what say ??

http://citeseerx.ist.psu.edu/viewdoc/... this is what i mean from scale expansion @berak

and motive is to compare current frame with that of previous and analyse, if the drone or vehicle is approaching towards an obstacle, its features diameter increases , i.e. the obstacle is getting bigger for the vehicle, and we need to avoid those points which are not growing in size or can not be considered as obstacle,, am i clear now ?

i am trying to do obstacke avoidance using SURF features detector

then if i could do this then all the features of a table,window , wall etc unless my vehicle is getting closer to it wont be considered as obstacle. . i hope i am clear

no i mean size of the feature by a factor of say 1.2 or some thing,, only those feature then i need to display whole size is changing , or the object is approaching near

hello guys, i am working on algorithm based on SURF feature detector, and i need to increase the size of the feature if the object in the frame is getting closer and ignore those features whose size is not changing...

so far i have this code :..

#include <iostream>
#include <stdio.h>
#include <opencv2/opencv.hpp>
#include <opencv2/core.hpp>
#include <opencv2/imgproc.hpp>
#include <opencv2/highgui.hpp>
#include <opencv2/videoio.hpp>
#include <opencv2/imgcodecs.hpp>
#include <opencv2/features2d.hpp>
#include <opencv2/features2d/features2d.hpp>
#include <opencv2/xfeatures2d.hpp>
#include <opencv2/calib3d.hpp>





using namespace cv;
using namespace std;
using namespace cv::xfeatures2d;

/** @function main */
int main( int argc, char** argv )
{


     VideoCapture cap(0);
     //Mat curr, prev;
     if(!cap.isOpened())return -1;
     Mat frame;
    Mat prevframe;

    while (cap.isOpened()) {
        cap.read(frame);


        // more code goes here which I haven't written here

        frame.copyTo(prevframe); // set previous frame to current frame
        //imshow("Video current", frame);
       // imshow("Videoprevious", frame);

        char key = waitKey(33);
        if (key == 'q')
        {
            break;
        }

  //-- Step 1: Detect the keypoints using SURF Detector
  int minHessian = 100;

  Ptr<SURF> detector = SURF::create( minHessian );
  detector->setHessianThreshold(minHessian);

  std::vector<KeyPoint> keypoints1,keypoints2;
  Mat descriptors_1, descriptors_2;

  detector->detectAndCompute( frame, Mat(), keypoints1, descriptors_1 );
  detector->detectAndCompute( prevframe, Mat(), keypoints2, descriptors_2 );

  //-- Step 2: Matching descriptor vectors using FLANN matcher
  FlannBasedMatcher matcher;
  std::vector< DMatch > matches;
  matcher.match( descriptors_1, descriptors_2, matches );
  double max_dist = 0; double min_dist = 100;

  //-- Quick calculation of max and min distances between keypoints
  for( int i = 0; i < descriptors_1.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 2*min_dist,
  //-- or a small arbitary value ( 0.02 ) in the event that min_dist is very
  //-- small)
  //-- PS.- radiusMatch can also be used here.


  std::vector< DMatch > good_matches;
  for( int i = 0; i < descriptors_1.rows; i++ )
  { if( matches[i].distance <= max(2*min_dist, 0.02) )
    { good_matches.push_back( matches[i]);}
  }
  //-- Draw only "good" matches
  Mat img_matches;
  drawMatches( frame, keypoints1, prevframe, keypoints2,
               good_matches, img_matches, Scalar::all(-1), Scalar::all(-1),
               vector<char>(), DrawMatchesFlags::NOT_DRAW_SINGLE_POINTS );

  //-- Show detected matches

  imshow( "Good Matches", img_matches );
  if( (good_matches.size() >=20)){

        for( unsigned int i = 0; i < good_matches.size(); i++ )
        {

        }
  }

    }
    waitKey(0);
    return 0;
}

please help

thanks all

its done berak, thanks

i tried adding opencv_flann it didnt work

http://answers.ros.org/question/26179... please help me out guys ..

thanks

it worked,, i had to do,, sudo apt-get update and sudo apt-get upgrade to resolve this , thanks @berak,, and sorry if i irritated you.. cheers

just final step :)

@berak this is working fine, and build finished successfully ,, thanks,, but i am not able to run ./tracker its showing me ./tracker: error while loading shared libraries: libopencv_tracking.so.3.2: cannot open shared object file: No such file or directory

https://pastebin.com/zfP8zJQJ

i did it,, now i am not getting any errors (thanks @berak) .. now its this https://pastebin.com/0kP3zzfi ,,

tried that too,, not happening

this is the code https://pastebin.com/bKi111q0 just giving error in create tracker

Building file: ../src/tracker2.cpp Invoking: GCC C++ Compiler g++ -I/usr/local/lib -I/usr/local/include -O0 -g3 -Wall -c -fmessage-length=0 -MMD -MP -MF"src/tracker2.d" -MT"src/tracker2.d" -o "src/tracker2.o" "../src/tracker2.cpp" ../src/tracker2.cpp: In function ‘int main(int, char)’: ../src/tracker2.cpp:14:44: error: expected primary-expression before ‘const’ Ptr<tracker> tracker = Tracker::create(const String& TrackerMIL ); ^ make: * [src/tracker2.o] Error 1 src/subdir.mk:18: recipe for target 'src/tracker2.o' failed

17:01:53 Build Finished (took 914ms)

this is not the include error !! some small thing is really missing ..

that was rude :(

how can i solve this

i solved the error now u«its just one thing if you can help @berak

Invoking: GCC C++ Linker g++ -L/usr/local/lib -L/usr/local/include -L/usr/include -o "tracker" ./src/tracker.o -lopencv_imgproc -lopencv_utility -lopencv_videoio -lopencv_highgui -lopencv_vedio -lopencv_core -lopencv_tracking /usr/bin/ld: cannot find -lopencv_utility /usr/bin/ld: cannot find -lopencv_vedio collect2: error: ld returned 1 exit status makefile:45: recipe for target 'tracker' failed make: * [tracker] Error 1

i got the flow,, i have opencv3 which came with my ros kinect , and i need to buikd my opencv_contrib there ,,, opencv3 after locating is in /opt/ros/kinetic/share/opencv3 ..

home/zubair/opencv_contrib/opencv_contrib/modules/tracking/include/opencv2/tracking/tracker.hpp:1341:56: error: template argument 1 is invalid CV_WRAP bool add(const Mat& image, std::vector<rect2d> boundingBox); ^ /home/zubair/opencv_contrib/opencv_contrib/modules/tracking/include/opencv2/tracking/tracker.hpp:1341:56: error: template argument 2 is invalid /home/zubair/opencv_contrib/opencv_contrib/modules/tracking/include/opencv2/tracking/tracker.hpp:1351:15: error: ‘Rect2d’ was not declared in this scope std::vector<rect2d> objects; ^

again, why am i getting errors in header file ? :(

what should i do next to rectify the error ?

yes berek so i have build all with opencv_contrib in opencv source folder

i am follwoing the instruction here http://docs.opencv.org/3.2.0/de/d25/t... , so far its building good,, i will be back after its done

where should the binaries be build ?

zubair@zubair-X541UV:~/opencv_contrib$ cmake -D CMAKE_BUILD_TYPE=Release -D CMAKE_INSTALL_PREFIX=/usr/local .. CMake Error: The source directory "/home/zubair" does not appear to contain CMakeLists.txt. Specify --help for usage, or press the help button on the CMake GUI. zubair@zubair-X541UV:~/opencv_contrib$

this is what i got after i follwed this link http://docs.opencv.org/master/d7/d9f/...

one thing berek,, in this link http://docs.opencv.org/master/d7/d9f/... ,, what should i add in place of cmake -D CMAKE_BUILD_TYPE=Release -D CMAKE_INSTALL_PREFIX=/usr/local .. what path ??

include for my opencv_contrib ?? you mean ?

how should i undo the extra module which i added in existing opencv ?'' and how to install then from build ? :(

zubair@zubair-X541UV:~$ g++ -I/usr/local/include g++: fatal error: no input files compilation terminated. zubair@zubair-X541UV:~$

it couldnt find

i dont have a build folder ,, just a make file :(

i actually just build the extra module .. from this link https://github.com/opencv/opencv_contrib ,,

actually is just the problem with the tracking.hpp,, something is going wrong ..

i did,, please check my comment .. i gave the whole error as pastebin link .