andrea's profile - activity

Hi to all! i try to use Orb feature for object detection... i use this code:

int numKeyPoints = 200;
std::vector<KeyPoint> keypoints_1,keypoints_2;
OrbFeatureDetector detector(numKeyPoints);
detector.detect( img_1, keypoints_1 );
detector.detect( img_2, keypoints_2 );
OrbDescriptorExtractor extractor;
Mat descriptors_1,descriptors_2;
extractor.compute( img_1, keypoints_1, descriptors_1 );
extractor.compute( img_2, keypoints_2, descriptors_2 );
BFMatcher matcher(Hamming);
std::vector< DMatch > matches;
matcher.match( descriptor_1, descriptor_2, matches );

but i have this error:

request for member 'match' in 'matcher', which is of non-class type 'cv::BFMatcher ()(cv::Hamming)'

I don't understand where is the error....

thanks!!

hi!

there are a method for usig warpperspective with the iplimage?

i try : warpPerspective(img_frame,crop,Homo.inv(DECOMP_LU),cvGetSize(img_temp));

where img_frame and crop are Iplimage* but i give me an error

error: invalid initialization of reference of type 'const cv::_InputArray&' from expression of type 'IplImage*' /opt/local/include/opencv2/imgproc/imgproc.hpp:540: error: in passing argument 1 of 'void cv::warpPerspective(const cv::_InputArray&, const cv::_OutputArray&, const cv::_InputArray&, cv::Size, int, int, const cv::Scalar&)'

yes...i tried this method but don't work...

I try to use VideoWriter::write... I create

VideoWriter writer2("video_rect_out2.avi", CV_FOURCC('M','J','P','G'), fps, cvSize(img_template.cols, img_template.rows), true);

and after i use

writer2 << img_template;

but at the end the video result is empty...

mmmm when i use this iplimage for the stream of the video tha video don't contain any image...

hi to all!

i have a simple question...can i convert a Mat imege into a Iplimage* ???

i need to insert the Mat image into a video,and the cvWriteFrema need a IplImage...

thanks :)

Hi to all!

I have a problem with the creation of a output video. In my program I take a video from the input and I recognize if there is some specific object (in this case the screen of a smartphone) and then I create the rectified image of the frame, so I have a new image with the video of the smartphone...

I must create the video with the image of the rectification of the screen of the smartphone... But at the end I have only a videofile of 8Kb... I think that the writer doesn't insert the image in the stream....

this is the definition of the writer:

// capture properties
frameH = (int) cvGetCaptureProperty(capture, CV_CAP_PROP_FRAME_HEIGHT);
frameW = (int) cvGetCaptureProperty(capture, CV_CAP_PROP_FRAME_WIDTH);
fps    = (int) cvGetCaptureProperty(capture, CV_CAP_PROP_FPS);
writer = cvCreateVideoWriter("video_rect_out.avi", CV_FOURCC('P','I','M','1'), fps, 
                             cvSize(frameW,frameH), isColor);

and this is the insert of the image in the stream (img_video_stream is the image of the rectification of the screen):

img_out = new IplImage(img_video_stream);
cvWriteFrame(writer, img_out);
cvShowImage("Rettificato out", img_out);

someone can have some idea of the problem?

i already try

i write: perspectiveTransform(keypoints[i],temp,inv)

but he give me an error : invalid initialization of reference of type 'const cv::_InputArray&' from expression of type 'cv::KeyPoint'

is it possible to apply a homography to the coordinates of a point?

for example in his code:

double X;
double Y;
std::vector<KeyPoint> key_filtered;
for( int i=0; i<(int)keypoints.size(); i++)
{
        double x = keypoints[i].pt.x, y = keypoints[i].pt.y;


        if((X >= corner[0].x && X <= corner[1].x) && (Y >= corner[0].y && Y <= corner[2].y)){
                key_filtered[j]=keypoints[i];
                j++;
        }
    }

i want apply the homography "inv" to "x" and "y"....

thanks!

can you give me an example of the use of warpPerspective()? i try warpPerspective(img_source,img_dest,H,img_dest.size)

but i give me an error

main.cpp:211: error: conversion from 'cv::Mat::MSize' to non-scalar type 'cv::Size' requested

hi to all!

i have a little problem...if i have a homography H and an image (call it img_source)....how can i apply this homography to the image? i try:

perspectiveTransform( img_source, img, H);

but when i run the program i give me an error....

thank a lot!

I read that for finding that the homography is good or bad thare are 2 solution:

1. Compute the determinant of the homography, and see if it's too close to zero for comfort.

2. Even better, compute its SVD, and verify that the ratio of the first-to-last singular value is sane (not too high). Either result will tell you whether the matrix is close to singular.

Is this true? and there is a method to calculate the SVD in opencv?

thanks!

i try to do some debug in my code and i have verify that the Homography H is wrong for some frames! the exact H is :

H = [0.6487030067242824, -0.08631426465278198, 261.6879493611227; 0.01335918943588019, 0.7381261562725441, 68.20078368844314; -0.0001412237402441972, -5.520086932186027e-06, 1]

and for some frames there is:

H = [-2.57857982218806, 0.3892895299010169, 394.0872195148755; -0.7676881270794974, 0.2898285736196523, 107.9234724910376; -0.006545151477964493, 0.0009948206740805412, 1]

now i tries with your method but i am not sure that i can change the SURF with the ORB....in this case some one know a method for correct the wrong Homography??????

thanks a lot!!!

sorry have another question... but if i want to crop a region that is not a rect? for example i have 4 point (the vertices) and i want the region delimited by this vertices (almost of the time is a rectangle but not always)

thanks :)

hi!

someone know how to use crop for the image?

for example i have one image i want to show a part of this image...

i read that there are the cvRect and cvSetImageROI.....

thanks a lot :)