
July 16th, 2011, 08:25 AM
#1
OpenCV and problems with matrices
Hello,
Ive been working on learning opencv by example and have learned lots but have some confusions/confusion in some parts, and would appreciate if you guys could help me out..
First off I have 5 matrices, and I want to combine them into one big matrix..
for this reason I converted them to row matrices, using the code:
Code:
CvMat row_header, *row, *row2;
row = cvReshape(mat, &row_header, 0, 1);
row2 = cvReshape(mat2, &row_header, 0, 1);
I then want to combine these into a single matrix, and for that I used:
Code:
cv::Mat M1(2,50,CV_32FC1);
M1.row(1) = mat + 0;
M1.row(2) = mat2 + 0;
IplImage imgc = M1;
CvMat *M2;
M2 = cvCreateMat(2,50,CV_32FC1 );
cvConvert(&imgc, M2);
cvSave( "combined.txt", M2 );
but this does not seem to work.. I tried different variations of the code and none of them work, wont even compile but with the code i wrote above it does compile but then gives some runtime errors, when I ignore that it continues and executes, but below is the saved matrix:
Code:
rows: 2
cols: 10
dt: f
data: [ 431602080., 431602080., 431602080., 431602080.,
431602080., 431602080., 431602080., 431602080., 431602080.,
431602080., 431602080., 431602080., 431602080., 431602080.,
431602080., 431602080., 431602080., 431602080., 431602080.,
431602080. ]
it does not look anything like the original matrices
help please.. how to i properly combine the matrices?
Thank you
P.S. mat and mat2 are just 2 simple matrices having 5 rows and 10 colums each.. im pretty sure my matrix combination method is at fault

July 16th, 2011, 11:01 AM
#2
Re: OpenCV and problems with matrices
Could you give an example demonstrating what you mean by "combined into one"?

July 16th, 2011, 04:42 PM
#3
Re: OpenCV and problems with matrices
suppose I have 2 matrices A and B..
A = [1,2,3,4,5]
B = [6,7,8,9,10]
I want to to combine them like so:
C = [1,2,3,4,5
6,7,8,9,10]
that each matrix is a row in the combined matrix
Regards

July 16th, 2011, 10:13 PM
#4
Re: OpenCV and problems with matrices
Update:
The strange runtime error is no more.. The problem was here:
I changed to M1.row(0) and M1.row(1), before it was giving error because I was telling the compiler that M1 has a third row as well with M1.row(2) and that is why i got the runtime error.
Code:
//now creating a combined matrix and copying these 2 into it
//CvMat *M1;
cv::Mat M1(2,row>width,CV_32FC1);
//M1 = cvCreateMat(2,50,CV_32FC1 );
//row.copyTo(M1.row(1));
//row2.copyTo(M1.row(2));
M1.row(0) = row + 0;
M1.row(1) = row2 + 0;
IplImage imgc = M1;
//CvMat matc = M1;
CvMat *M2;
M2 = cvCreateMat(2,row>width,CV_32FC1 );
cvConvert(&imgc, M2);
//cvConvert(M1, M2 );
cvSave( "combined.txt", M2 );
this fixed the strange runtime error I used to get, and the program executes without any problems.. but the matrix and combined.txt is still filled with junk values, here is the combined.txt after the above changes:
Code:
rows: 2
cols: 50
dt: f
data: [ 431602080., 431602080., 431602080., 431602080.,
431602080., 431602080., 431602080., 431602080., 431602080.,
431602080., 431602080., 431602080., 431602080., 431602080.,
431602080., 431602080., 431602080., 431602080., 431602080.,
431602080., 431602080., 431602080., 431602080., 431602080.,
431602080., 431602080., 431602080., 431602080., 431602080.,
431602080., 431602080., 431602080., 431602080., 431602080.,
431602080., 431602080., 431602080., 431602080., 431602080.,
431602080., 431602080., 431602080., 431602080., 431602080.,
431602080., 431602080., 431602080., 431602080., 431602080.,
431602080., 431602080., 431602080., 431602080., 431602080.,
431602080., 431602080., 431602080., 431602080., 431602080.,
431602080., 431602080., 431602080., 431602080., 431602080.,
431602080., 431602080., 431602080., 431602080., 431602080.,
431602080., 431602080., 431602080., 431602080., 431602080.,
431602080., 431602080., 431602080., 431602080., 431602080.,
431602080., 431602080., 431602080., 431602080., 431602080.,
431602080., 431602080., 431602080., 431602080., 431602080.,
431602080., 431602080., 431602080., 431602080., 431602080.,
431602080., 431602080., 431602080., 431602080., 431602080.,
431602080. ]
just for reference.. when the files transferred to the combined matrix rows are exported they show actual values and not junk.. here is the exported txt file of the matrix "row":
Code:
rows: 1
cols: 50
dt: f
data: [ 60., 89., 86., 102., 58., 51., 143., 187., 140., 64., 80.,
169., 184., 172., 67., 90., 174., 191., 175., 41., 97., 86., 171.,
104., 87., 145., 164., 176., 157., 119., 176., 177., 180., 179.,
150., 44., 182., 148., 182., 38., 48., 197., 171., 171., 39., 48.,
169., 163., 141., 39. ]
I checked my code again and could not find out as to why the combined matrix is filled with junk values.. help here would be appreciated

July 19th, 2011, 11:58 AM
#5
Re: OpenCV and problems with matrices
just wanted to update here for people having similar problems.. I fixed it with the following code:
Code:
CvMat *M1;
//cv::Mat M1(2,row>width,CV_32FC1);
M1 = cvCreateMat(5,row>width,CV_32FC1 );
//row.copyTo(M1.row(0));
//row2.copyTo(M1.row(2));
//M1.row(0) = 0;
//M1.row(0) = row + 0;
//M1.row(1) = 0;
//M1.row(1) = row2 + 0;
cvSetZero(M1);
/*for(int i=0;i<5;i++)
{
for(int j=0;j<row>width;j++)
{
CV_MAT_ELEM( *M1, float, i, j ) = CV_MAT_ELEM( *rowi, float, 0, i );*/
for(int i=0;i<row>width;i++)
{
CV_MAT_ELEM( *M1, float, 0, i ) = CV_MAT_ELEM( *row, float, 0, i );
}
for(int j=0;j<row2>width;j++)
{
CV_MAT_ELEM( *M1, float, 1, j ) = CV_MAT_ELEM( *row2, float, 0, j );
}
for(int k=0;k<row2>width;k++)
{
CV_MAT_ELEM( *M1, float, 2, k ) = CV_MAT_ELEM( *row3, float, 0, k );
}
for(int l=0;l<row2>width;l++)
{
CV_MAT_ELEM( *M1, float, 3, l ) = CV_MAT_ELEM( *row3, float, 0, l );
}
for(int m=0;m<row2>width;m++)
{
CV_MAT_ELEM( *M1, float, 4, m ) = CV_MAT_ELEM( *row4, float, 0, m );
}
the matrix M1 is a 5 row matrix having 50 columns.. each row represents a seperate matrix.. row, row2, row3, row4 and row5
Tags for this Thread
Posting Permissions
 You may not post new threads
 You may not post replies
 You may not post attachments
 You may not edit your posts

Forum Rules

Click Here to Expand Forum to Full Width
OnDemand Webinars (sponsored)
