I have created a 3D matrix in opencv with the size: 120x100x50 :
int _sz[] = {120,100,50};
Mat src(3,_sz,CV_32FC1,Scalar(2));
I want to transpose the first and the second dimension of this matrix to 100x120x50.
I use 3 loop iteration to manually transpose
for ()
for ()
for ()
dst.at<float>( , , ) = src.at<float>( , , );
But it takes too long with this way. And, I cannot utilize channel as the third dimension because the third dimension is 50 whereas OpenCV mat only supports 4 channels in maximum. Are there any other way to do this work quicker? Just do something like OpenCV did with multichannel matrix because when I test with matrix 120x100x4 by using 4 channel as the third dimension, the execution time is much better.
I change to access values of matrix like array values. Execution time is much better. But it still accesses one-by-one, not copy block-by-block.
int indexOrder[3];
int order[3]={1,0,2};
int* _srcSize = srcMat.size.p;
int srcMatSize[] = {_srcSize[0],_srcSize[1],_srcSize[2]};
int dstMatSize[] = {srcMatSize[order[0]], srcMatSize[order[1]], srcMatSize[order[2]]};
Mat dstMat(3, dstMatSize, CV_32FC1);
float* srcArr, *dstArr;
srcArr = srcMat.ptr<float>();
dstArr = dstMat.ptr<float>();
for (indexOrder[0]=0;indexOrder[0]<srcMatSize[0];indexOrder[0]++)
for (indexOrder[1]=0;indexOrder[1]<srcMatSize[1];indexOrder[1]++)
for (indexOrder[2]=0;indexOrder[2]<srcMatSize[2];indexOrder[2]++)
dstArr[indexOrder[order[0]]*dstMatSize[1] + indexOrder[order[1]]*dstMatSize[2] + indexOrder[order[2]]] =
srcArr[indexOrder[0]*srcMatSize[1]+indexOrder[1]*srcMatSize[2]+indexOrder[2]];
Related
I am using OpenCV 3.1 with VS2012 C++/CLI.
I have stored the result of a finContours call into:
std::vector<std::vector<Point>> Contours;
Thus, Contours[0] is a vector of the contour points of the first contour.
Contours[1] is a vector of the contour points of the second vector, etc.
Now, I want to load one of the contours into a Mat Based on Convert Mat to vector <float> and Vector<float> to mat in opencv I thought something like this would work.
Mat testMat=Mat(Images->Contours[0].size(),2,CV_32FC1);
memcpy(testMat.data,Images->Contours[0].data(),Images->Contours[0].size()*CV_32FC1);
I specified two columns because I each underlying pint must be composed of both an X point and a Y point and each of those should be a float. However, when I access the Mat elements, I can see that the first element is not the underlying data but the total number of contour points.
Any help on the right way to accomplish this appreaciated.
You can do that with:
Mat testMat = Mat(Images->Contours[0]).reshape(1);
Now testMat is of type CV_32SC1, aka of int. If you need float you can:
testMat.convertTo(testMat, CV_32F);
Some more details and variants...
You can simply use the Mat constructor that accepts a std::vector:
vector<Point> v = { {0,1}, {2,3}, {4,5} };
Mat m(v);
With this, you get a 2 channel matrix with the underlying data in v. This means that if you change the value in v, also the values in m change.
v[0].x = 7; // also 'm' changes
If you want a deep copy of the values, so that changes in v are not reflected in m, you can use clone:
Mat m2 = Mat(v).clone();
Your matrices are of type CV_32SC2, i.e. 2 channels matrices of int (because Point uses int. Use Point2f for float). If you want a 2 columns single channel matrix you can use reshape:
Mat m3 = m2.reshape(1);
If you want to convert to float type, you need to use convertTo:
Mat m4;
m2.convertTo(m4, CV_32F);
Here some working code as a proof of concept:
#include <opencv2\opencv.hpp>
#include <vector>
using namespace std;
using namespace cv;
int main()
{
vector<Point> v = { {0,1}, {2,3}, {4,5} };
// changes in v affects m
Mat m(v);
// changes in v doesn't affect m2
Mat m2 = Mat(v).clone();
// m is changed
v[0].x = 7;
// m3 is a 2 columns single channel matrix
Mat m3 = m2.reshape(1);
// m4 is a matrix of floats
Mat m4;
m2.convertTo(m4, CV_32F);
return 0;
}
I am trying to port an existing FFT based low-pass filter to iOS using the Accelerate vDSP framework.
It seems like the FFT works as expected for about the first 1/4 of the sample. But then after that the results seem wrong, and even more odd are mirrored (with the last half of the signal mirroring most of the first half).
You can see the results from a test application below. First is plotted the original sampled data, then an example of the expected filtered results (filtering out signal higher than 15Hz), then finally the results of my current FFT code (note that the desired results and example FFT result are at a different scale than the original data):
The actual code for my low-pass filter is as follows:
double *lowpassFilterVector(double *accell, uint32_t sampleCount, double lowPassFreq, double sampleRate )
{
double stride = 1;
int ln = log2f(sampleCount);
int n = 1 << ln;
// So that we get an FFT of the whole data set, we pad out the array to the next highest power of 2.
int fullPadN = n * 2;
double *padAccell = malloc(sizeof(double) * fullPadN);
memset(padAccell, 0, sizeof(double) * fullPadN);
memcpy(padAccell, accell, sizeof(double) * sampleCount);
ln = log2f(fullPadN);
n = 1 << ln;
int nOver2 = n/2;
DSPDoubleSplitComplex A;
A.realp = (double *)malloc(sizeof(double) * nOver2);
A.imagp = (double *)malloc(sizeof(double) * nOver2);
// This can be reused, just including it here for simplicity.
FFTSetupD setupReal = vDSP_create_fftsetupD(ln, FFT_RADIX2);
vDSP_ctozD((DSPDoubleComplex*)padAccell,2,&A,1,nOver2);
// Use the FFT to get frequency counts
vDSP_fft_zripD(setupReal, &A, stride, ln, FFT_FORWARD);
const double factor = 0.5f;
vDSP_vsmulD(A.realp, 1, &factor, A.realp, 1, nOver2);
vDSP_vsmulD(A.imagp, 1, &factor, A.imagp, 1, nOver2);
A.realp[nOver2] = A.imagp[0];
A.imagp[0] = 0.0f;
A.imagp[nOver2] = 0.0f;
// Set frequencies above target to 0.
// This tells us which bin the frequencies over the minimum desired correspond to
NSInteger binLocation = (lowPassFreq * n) / sampleRate;
// We add 2 because bin 0 holds special FFT meta data, so bins really start at "1" - and we want to filter out anything OVER the target frequency
for ( NSInteger i = binLocation+2; i < nOver2; i++ )
{
A.realp[i] = 0;
}
// Clear out all imaginary parts
bzero(A.imagp, (nOver2) * sizeof(double));
//A.imagp[0] = A.realp[nOver2];
// Now shift back all of the values
vDSP_fft_zripD(setupReal, &A, stride, ln, FFT_INVERSE);
double *filteredAccell = (double *)malloc(sizeof(double) * fullPadN);
// Converts complex vector back into 2D array
vDSP_ztocD(&A, stride, (DSPDoubleComplex*)filteredAccell, 2, nOver2);
// Have to scale results to account for Apple's FFT library algorithm, see:
// http://developer.apple.com/library/ios/#documentation/Performance/Conceptual/vDSP_Programming_Guide/UsingFourierTransforms/UsingFourierTransforms.html#//apple_ref/doc/uid/TP40005147-CH202-15952
double scale = (float)1.0f / fullPadN;//(2.0f * (float)n);
vDSP_vsmulD(filteredAccell, 1, &scale, filteredAccell, 1, fullPadN);
// Tracks results of conversion
printf("\nInput & output:\n");
for (int k = 0; k < sampleCount; k++)
{
printf("%3d\t%6.2f\t%6.2f\t%6.2f\n", k, accell[k], padAccell[k], filteredAccell[k]);
}
// Acceleration data will be replaced in-place.
return filteredAccell;
}
In the original code the library was handling non power-of-two sizes of input data; in my Accelerate code I am padding out the input to the nearest power of two. In the case of the sample test below the original sample data is 1000 samples so it's padded to 1024. I don't think that would affect results but I include that for the sake of possible differences.
If you want to experiment with a solution, you can download the sample project that generates the graphs here (in the FFTTest folder):
FFT Example Project code
Thanks for any insight, I've not worked with FFT's before so I feel like I am missing something critical.
If you want a strictly real (not complex) result, then the data before the IFFT must be conjugate symmetric. If you don't want the result to be mirror symmetric, then don't zero the imaginary component before the IFFT. Merely zeroing bins before the IFFT creates a filter with a huge amount of ripple in the passband.
The Accelerate framework also supports more FFT lengths than just powers of 2.
I am fairly new to OpenCV and sort of understanding it bit by bit. I know that the matrix operators in cv::Mat class has been overloaded to do A.mult(B), A+B, A-B, A/B, etc.
I have two vectors which are projections of rows and columns of an image. I have two images(S and T), so each of them will have two projection vectors (rowProejctionS, columnProjectionS, rowProjectionT, columnProjectionT). I also have the means of the images (meanS, meanT). I need to do a "SUM OF PRODUCT" related calculation, which in MATLAB is as follows
numeratorLambdaRo = sum((rowProjectionT - meanT).*(rowProjectionS - meanS));
denominatorLambdaRo = sqrt(sum((rowProjectionT - meanT).^2)*sum((rowProjectionS - meanS).^2);
LambaRo = numeratorLambdaRo/denominatorLambdaRo;
I am not entirely sure about the capability of matrix operators in the context of cv::Mat objects.
declare meanT, meanS as double or cv::Scalar and you can just substract it from your matrix. You can maybe split your operations :
rowProjectionT -= meanT;
rowProjectionS -= meanS;
numeratoLambdaRo = cv::sum(rowProjectionT*rowProjectionS.t()); // transpose 1 of the vector so that multiplication is equivalent to dot product.
cv::Mat rowProjTSquare = rowProjectionT*rowProjectionT.t();
cv::Mat rowProjSSquare = rowProjectionS*rowProjectionS.t();
denominatorLambdaRo = sqrt(cv::sum(rowProjTSquare*rowProjSSquare));
Since each pixel memory contains 8 bit for each component Blue,Green and Red. So how can I separate these components from Image or Image Matrix. As
int Blue = f(Image(X,y));// (x,y) = Coordinate of a pixel of Image
similarly, for red and green.
So what should be function f and 2D matrix Image;
Thanks in advance
First off, you must go through the basics of OpenCV and turn your attention towards other parts of image processing. What you ask for is pretty basic and assuming you will be using OpenCV 2.1 and higher,
cv::Mat img = Read the image off the disk or do something to fill the image.
To access the RGB values
img.at<cv::Vec3b>(x,y);
But would give the values in reverse that is BGR. So make sure you note this.
Basically a cv::Vec3b type that is accessed.
img.at<cv::Vec3b>(x,y)[0];//B
img.at<cv::Vec3b>(x,y)[1];//G
img.at<cv::Vec3b>(x,y)[2];//R
or
Vec3f pixel = img.at<Vec3f>(x, y);
int b = pixel[0];
int g = pixel[1];
int r = pixel[2];
Now onto splitting the image into RGB channels you can use the following
Now down to primitive C style of OpenCV (There C and C++ style supported)
You can use the cvSplit function
IplImage* rgb = cvLoatImage("C://MyImage.bmp");
//now create three single channel images for the channel separation
IplImage* r = cvCreateImage( cvGetSize(rgb), rgb->depth,1 );
IplImage* g = cvCreateImage( cvGetSize(rgb), rgb->depth,1 );
IplImage* b = cvCreateImage( cvGetSize(rgb), rgb->depth,1 );
cvSplit(rgb,b,g,r,NULL);
OpenCV 2 CookBook Is one of the best books on OpenCV. Will help you alot.
I'm trying to create a PCA model in OpenCV to hold pixel coordinates. As an experiment I have two sets of pixel coordinates that maps out two approximate circles. Each set of coordiantes has 48 x,y pairs. I was experimenting with the following code which reads the coordinates from a file and stores them in a Mat structure. However, I don't think it is right and PCA in openCV seems very poorly covered on the Internet.
Mat m(2, 48, CV_32FC2); // matrix with 2 rows of 48 cols of floats held in two channels
pFile = fopen("data.txt", "r");
for (int i=0; i<48; i++){
int x, y;
fscanf(pFile, "%d%c%c%d%c", &x, &c, &c, &y, &c);
m.at<Vec2f>( 0 , i )[0] = (float)x; // store x in row 0, col i in channel 0
m.at<Vec2f>( 0 , i )[1] = (float)y; // store y in row 0, col i in channel 1
}
for (int i=0; i<48; i++){
int x, y;
fscanf(pFile, "%d%c%c%d%c", &x, &c, &c, &y, &c);
m.at<Vec2f>( 1 , i )[0] = (float)x; // store x in row 1, col i in channel 0
m.at<Vec2f>( 1 , i )[1] = (float)y; // store y in row 1, col i in channel 1
}
PCA pca(m, Mat(), CV_PCA_DATA_AS_ROW, 2); // 2 principle components??? Not sure what to put here e.g. is it 2 for two data sets or 48 for number of elements?
for (int i=0; i<48; i++){
float x = pca.mean.at<Vec2f>(i,0)[0]; //get average x
float y = pca.mean.at<Vec2f>(i,0)[1]; //get average y
printf("\n x=%f, y=%f", x, y);
}
However, this crashes when creating the pca object. I know this is a very basic question but I am a bit lost and was hoping that someone could get me started with pca in open cv.
Perhaps it would be helpful if you described in further detail what you need to use PCA for and what you hope to achieve (output?).
I am fairly sure that the reason your program crashes is because the input Mat is CV_32FC2, when it should be CV_32FC1. You need to reshape your data into 1 dimensional row vectors before using PCA, not knowing what you need I can't say how to reshape your data. (The common application with images is eigenFace which requires an image to be reshaped into a row vector). Additionally you will need to normalize your input data between 0 and 1.
As a further aside, usually you would choose to keep 1 less principal component than the number of input samples because the last principal component is simply orthogonal to the others.
I have worked with opencv PCA before and would like to help further. I would also refer you to this blog: http://www.bytefish.de/blog/pca_in_opencv which helped me get started with PCA in openCV.