Matchers of keypoint descriptors in OpenCV have wrappers with common interface that enables to switch easily between different algorithms solving the same problem. This section is devoted to matching descriptors that are represented as vectors in a multidimensional space. All objects that implement ‘’vector’’ descriptor matchers inherit DescriptorMatcher() interface.
Match between two keypoint descriptors: query descriptor index, train descriptor index, train image index and distance between descriptors.
struct DMatch
{
DMatch() : queryIdx(-1), trainIdx(-1), imgIdx(-1),
distance(std::numeric_limits<float>::max()) {}
DMatch( int _queryIdx, int _trainIdx, float _distance ) :
queryIdx(_queryIdx), trainIdx(_trainIdx), imgIdx(-1),
distance(_distance) {}
DMatch( int _queryIdx, int _trainIdx, int _imgIdx, float _distance ) :
queryIdx(_queryIdx), trainIdx(_trainIdx), imgIdx(_imgIdx),
distance(_distance) {}
int queryIdx; // query descriptor index
int trainIdx; // train descriptor index
int imgIdx; // train image index
float distance;
// less is better
bool operator<( const DMatch &m ) const;
};
Abstract base class for matching keypoint descriptors. It has two groups of match methods: for matching descriptors of one image with other image or with image set.
class DescriptorMatcher
{
public:
virtual ~DescriptorMatcher();
virtual void add( const vector<Mat>& descriptors );
const vector<Mat>& getTrainDescriptors() const;
virtual void clear();
bool empty() const;
virtual bool isMaskSupported() const = 0;
virtual void train();
/*
* Group of methods to match descriptors from image pair.
*/
void match( const Mat& queryDescriptors, const Mat& trainDescriptors,
vector<DMatch>& matches, const Mat& mask=Mat() ) const;
void knnMatch( const Mat& queryDescriptors, const Mat& trainDescriptors,
vector<vector<DMatch> >& matches, int k,
const Mat& mask=Mat(), bool compactResult=false ) const;
void radiusMatch( const Mat& queryDescriptors, const Mat& trainDescriptors,
vector<vector<DMatch> >& matches, float maxDistance,
const Mat& mask=Mat(), bool compactResult=false ) const;
/*
* Group of methods to match descriptors from one image to image set.
*/
void match( const Mat& queryDescriptors, vector<DMatch>& matches,
const vector<Mat>& masks=vector<Mat>() );
void knnMatch( const Mat& queryDescriptors, vector<vector<DMatch> >& matches,
int k, const vector<Mat>& masks=vector<Mat>(),
bool compactResult=false );
void radiusMatch( const Mat& queryDescriptors, vector<vector<DMatch> >& matches,
float maxDistance, const vector<Mat>& masks=vector<Mat>(),
bool compactResult=false );
virtual void read( const FileNode& );
virtual void write( FileStorage& ) const;
virtual Ptr<DescriptorMatcher> clone( bool emptyTrainData=false ) const = 0;
static Ptr<DescriptorMatcher> create( const string& descriptorMatcherType );
protected:
vector<Mat> trainDescCollection;
...
};
Add descriptors to train descriptor collection. If collection trainDescCollectionis not empty
the new descriptors are added to existing train descriptors.
param descriptors: Descriptors to add. Each descriptors[i] is a set of descriptors from the same (one) train image.
Return true if there are not train descriptors in collection.
Returns true if descriptor matcher supports masking permissible matches.
Train descriptor matcher (e.g. train flann index). In all methods to match the method train()
is run every time before matching. Some descriptor matchers (e.g. BruteForceMatcher) have empty implementation of this method, other matchers realy train their inner structures (e.g. FlannBasedMatcher trains flann::Index)
Find the best match for each descriptor from a query set with train descriptors.
Supposed that the query descriptors are of keypoints detected on the same query image. In first variant of this method train descriptors are set as input argument and supposed that they are of keypoints detected on the same train image. In second variant of the method train descriptors collection that was set using addmethod is used. Optional mask (or masks) can be set to describe which descriptors can be matched. queryDescriptors[i]can be matched with trainDescriptors[j]only if mask.at<uchar>(i,j)is non-zero.
Parameters: |
|
---|
DescriptorMatcher::match()
Find the k best matches for each descriptor from a query set with train descriptors.
Found k (or less if not possible) matches are returned in distance increasing order. Details about query and train descriptors see in .
Parameters: |
|
---|
DescriptorMatcher::match()
Find the best matches for each query descriptor which have distance less than given threshold.
Found matches are returned in distance increasing order. Details about query and train descriptors see in .
Parameters: |
|
---|
Clone the matcher.
Parameters: |
|
---|
Brute-force descriptor matcher. For each descriptor in the first set, this matcher finds the closest descriptor in the second set by trying each one. This descriptor matcher supports masking permissible matches between descriptor sets.
template<class Distance>
class BruteForceMatcher : public DescriptorMatcher
{
public:
BruteForceMatcher( Distance d = Distance() );
virtual ~BruteForceMatcher();
virtual bool isMaskSupported() const;
virtual Ptr<DescriptorMatcher> clone( bool emptyTrainData=false ) const;
protected:
...
}
For efficiency, BruteForceMatcher is templated on the distance metric. For float descriptors, a common choice would be L2<float> . Class of supported distances are:
template<typename T>
struct Accumulator
{
typedef T Type;
};
template<> struct Accumulator<unsigned char> { typedef unsigned int Type; };
template<> struct Accumulator<unsigned short> { typedef unsigned int Type; };
template<> struct Accumulator<char> { typedef int Type; };
template<> struct Accumulator<short> { typedef int Type; };
/*
* Squared Euclidean distance functor
*/
template<class T>
struct L2
{
typedef T ValueType;
typedef typename Accumulator<T>::Type ResultType;
ResultType operator()( const T* a, const T* b, int size ) const;
};
/*
* Manhattan distance (city block distance) functor
*/
template<class T>
struct CV_EXPORTS L1
{
typedef T ValueType;
typedef typename Accumulator<T>::Type ResultType;
ResultType operator()( const T* a, const T* b, int size ) const;
...
};
/*
* Hamming distance (city block distance) functor
*/
struct HammingLUT
{
typedef unsigned char ValueType;
typedef int ResultType;
ResultType operator()( const unsigned char* a, const unsigned char* b,
int size ) const;
...
};
struct Hamming
{
typedef unsigned char ValueType;
typedef int ResultType;
ResultType operator()( const unsigned char* a, const unsigned char* b,
int size ) const;
...
};
Flann based descriptor matcher. This matcher trains flann::Index() on train descriptor collection and calls it’s nearest search methods to find best matches. So this matcher may be faster in cases of matching to large train collection than brute force matcher. FlannBasedMatcher does not support masking permissible matches between descriptor sets, because flann::Index() does not support this.
class FlannBasedMatcher : public DescriptorMatcher
{
public:
FlannBasedMatcher(
const Ptr<flann::IndexParams>& indexParams=new flann::KDTreeIndexParams(),
const Ptr<flann::SearchParams>& searchParams=new flann::SearchParams() );
virtual void add( const vector<Mat>& descriptors );
virtual void clear();
virtual void train();
virtual bool isMaskSupported() const;
virtual Ptr<DescriptorMatcher> clone( bool emptyTrainData=false ) const;
protected:
...
};