CvMat, CvSeq and CvArr Data Structure
CvMat is multi-dimensional array structure, CvSeq a sequence of points, CvArr is their super structure.
- CvArr is used only a function parameter to specify that the function accepts arrays of more than a single type, for example IplImage*, CvMat * or even CvSeq *.
- CvMat Multi-dimensional matrix
Example: Calculating Product of Two Matrices
double a[] = { 1, 2, 3, 4
5, 6, 7, 8,
9, 10, 11, 12 };
double b[] = { 1, 5, 9,
2, 6, 10,
3, 7, 11,
4, 8, 12 };
double c[9];
CvMat Ma, Mb, Mc ;
cvInitMatHeader( &Ma, 3, 4, CV_64FC1, a );
cvInitMatHeader( &Mb, 4, 3, CV_64FC1, b );
cvInitMatHeader( &Mc, 3, 3, CV_64FC1, c );
cvMatMulAdd( &Ma, &Mb, 0, &Mc );
// c array now contains product of a(3x4) and b(4x3) matrices
- CvSeq is a base for all of OpenCV dynamic data structures.
functions:
- CreateSeq?: create sequence
- SeqPush?, SeqPushFront?, SeqPushMulti?: adds element to sequence
- SeqPop?, SeqPopFront?, SeqPopMulti?: removes element from sequence end
- SeqInsert?: inserts element in sequence middle
- SeqElemIdx?: search for an element in the sequence, if found returns the index, if not return -1
- CvtSeqToArray?: copies sequence to one continous block of memory
- SeqSort?: Sorts sequence element using the specified comparison function
Compute Bounding Boxes:
CvRect cvBoundingRect( CvArr* points, int update=0)
points: 2D point set, either a sequence (CvContour, CvSeq) or vector (CvMat) of points
Example:
CvSeq *contour;
double area;
CvRect rect;
cvFindContours(img, storage, &contour, size(CvContour),
CV_RETR_LIST, CV_CHAIN_APPROX_SIMPLE, cvPoint(0,0));
//go through each contour in the contours and compute their area as well as
//bounding box
while(contours!=NULL){
area = fabs(cvContourArea(contour, CV_WHOLE));
rect = cvBoundingRect(contour);
cout << "area: " << area << ", bounding box width/height: " << rect.width <<
"/" << rect.height << endl;
contour = contour->h_next; //get next contour
}