Learning OpenCV, 2nd Edition

int ksize, anchor;

};

class cv::BaseColumnFilter {

public:

virtual ~BaseColumnFilter();

// To be overriden by the user.

//

// runs a filtering operation on the set of rows,

// "dstcount + ksize - 1" rows on input,

// "dstcount" rows on output,

// each input and output row has "width" elements

// the filtered rows are written into "dst" buffer.

//

virtual void operator()(

const uchar** src,

uchar* dst,

int dststep,

int dstcount,

int width

) = 0;

// resets the filter state (may be needed for IIR filters)

//

virtual void reset();

int ksize, anchor;

};

The cv::BaseRowFilter class is the simplest because row data is sequential in memory, and so there

is a little less bookkeeping. The associated cv::BaseRowFilter::operator() inherits this relative

simplicity. The arguments are just the pointers to the source and destination rows, the width of the rows,

and the number of channels, cn. As with cv::BaseFilter::operator(), the source and destination

pointers are of type uchar*, and so it will be your responsibility to cast them to the correct type.

The cv::BaseColumnFilter class needs to handle the non-sequential nature of the required data. As a

result, there are the additional arguments dststep and dstcount, which the row filter did not require

28

.

These have the same meaning as they did for the cv::BaseFilter() class.

Base Filter Builders

There will