mirror/opencv
1
0
Fork 0
mirror of https://github.com/opencv/opencv.git synced 2024-11-29 13:47:32 +08:00
Code Issues Actions Packages Projects Releases Wiki Activity

Merge remote-tracking branch 'itseezstream/master'

Browse Source
This commit is contained in:
Ozan Tonkal 2013-09-17 17:50:14 +02:00
commit e3e5fd5baa
3 changed files with 63 additions and 49 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

28
modules/objdetect/include/opencv2/objdetect/erfilter.hpp
View File

@ -47,6 +47,7 @@
#include "opencv2/core.hpp"
#include <vector>
#include <deque>
#include <string>
namespace cv
{
@ -163,7 +164,8 @@ public:
local minimum is greater than minProbabilityDiff).
\param cb Callback with the classifier.
if omitted tries to load a default classifier from file trained_classifierNM1.xml
default classifier can be implicitly load with function loadClassifierNM1()
from file in samples/cpp/trained_classifierNM1.xml
\param thresholdDelta Threshold step in subsequent thresholds when extracting the component tree
\param minArea The minimum area (% of image size) allowed for retreived ER's
\param minArea The maximum area (% of image size) allowed for retreived ER's
@ -171,7 +173,7 @@ public:
\param nonMaxSuppression Whenever non-maximum suppression is done over the branch probabilities
\param minProbability The minimum probability difference between local maxima and local minima ERs
*/
CV_EXPORTS Ptr<ERFilter> createERFilterNM1(const Ptr<ERFilter::Callback>& cb = Ptr<ERFilter::Callback>(),
CV_EXPORTS Ptr<ERFilter> createERFilterNM1(const Ptr<ERFilter::Callback>& cb,
int thresholdDelta = 1, float minArea = 0.00025,
float maxArea = 0.13, float minProbability = 0.4,
bool nonMaxSuppression = true,
@ -187,13 +189,31 @@ CV_EXPORTS Ptr<ERFilter> createERFilterNM1(const Ptr<ERFilter::Callback>& cb = P
additional features: hole area ratio, convex hull ratio, and number of outer inflexion points.
\param cb Callback with the classifier
if omitted tries to load a default classifier from file trained_classifierNM2.xml
default classifier can be implicitly load with function loadClassifierNM2()
from file in samples/cpp/trained_classifierNM2.xml
\param minProbability The minimum probability P(er|character) allowed for retreived ER's
*/
CV_EXPORTS Ptr<ERFilter> createERFilterNM2(const Ptr<ERFilter::Callback>& cb = Ptr<ERFilter::Callback>(),
CV_EXPORTS Ptr<ERFilter> createERFilterNM2(const Ptr<ERFilter::Callback>& cb,
float minProbability = 0.3);
/*!
Allow to implicitly load the default classifier when creating an ERFilter object.
The function takes as parameter the XML or YAML file with the classifier model
(e.g. trained_classifierNM1.xml) returns a pointer to ERFilter::Callback.
*/
CV_EXPORTS Ptr<ERFilter::Callback> loadClassifierNM1(const std::string& filename);
/*!
Allow to implicitly load the default classifier when creating an ERFilter object.
The function takes as parameter the XML or YAML file with the classifier model
(e.g. trained_classifierNM1.xml) returns a pointer to ERFilter::Callback.
*/
CV_EXPORTS Ptr<ERFilter::Callback> loadClassifierNM2(const std::string& filename);
// computeNMChannels operation modes
enum { ERFILTER_NM_RGBLGrad = 0,
ERFILTER_NM_IHSGrad = 1

76
modules/objdetect/src/erfilter.cpp
View File

@ -137,7 +137,7 @@ class CV_EXPORTS ERClassifierNM1 : public ERFilter::Callback
{
public:
//Constructor
ERClassifierNM1();
ERClassifierNM1(const std::string& filename);
// Destructor
~ERClassifierNM1() {};
@ -153,7 +153,7 @@ class CV_EXPORTS ERClassifierNM2 : public ERFilter::Callback
{
public:
//constructor
ERClassifierNM2();
ERClassifierNM2(const std::string& filename);
// Destructor
~ERClassifierNM2() {};
@ -988,24 +988,13 @@ int ERFilterNM::getNumRejected()
// load default 1st stage classifier if found
ERClassifierNM1::ERClassifierNM1()
ERClassifierNM1::ERClassifierNM1(const std::string& filename)
{
if (ifstream("./trained_classifierNM1.xml"))
{
// The file with default classifier exists
boost.load("./trained_classifierNM1.xml", "boost");
}
else if (ifstream("./training/trained_classifierNM1.xml"))
{
// The file with default classifier exists
boost.load("./training/trained_classifierNM1.xml", "boost");
}
if (ifstream(filename.c_str()))
boost.load( filename.c_str(), "boost" );
else
{
// File not found
CV_Error(CV_StsBadArg, "Default classifier ./trained_classifierNM1.xml not found!");
}
CV_Error(CV_StsBadArg, "Default classifier file not found!");
};
double ERClassifierNM1::eval(const ERStat& stat)
@ -1026,24 +1015,12 @@ double ERClassifierNM1::eval(const ERStat& stat)
// load default 2nd stage classifier if found
ERClassifierNM2::ERClassifierNM2()
ERClassifierNM2::ERClassifierNM2(const std::string& filename)
{
if (ifstream("./trained_classifierNM2.xml"))
{
// The file with default classifier exists
boost.load("./trained_classifierNM2.xml", "boost");
}
else if (ifstream("./training/trained_classifierNM2.xml"))
{
// The file with default classifier exists
boost.load("./training/trained_classifierNM2.xml", "boost");
}
if (ifstream(filename.c_str()))
boost.load( filename.c_str(), "boost" );
else
{
// File not found
CV_Error(CV_StsBadArg, "Default classifier ./trained_classifierNM2.xml not found!");
}
CV_Error(CV_StsBadArg, "Default classifier file not found!");
};
double ERClassifierNM2::eval(const ERStat& stat)
@ -1079,7 +1056,8 @@ double ERClassifierNM2::eval(const ERStat& stat)
local minimum is greater than minProbabilityDiff).
\param cb Callback with the classifier.
if omitted tries to load a default classifier from file trained_classifierNM1.xml
default classifier can be implicitly load with function loadClassifierNM1()
from file in samples/cpp/trained_classifierNM1.xml
\param thresholdDelta Threshold step in subsequent thresholds when extracting the component tree
\param minArea The minimum area (% of image size) allowed for retreived ER's
\param minArea The maximum area (% of image size) allowed for retreived ER's
@ -1099,9 +1077,6 @@ Ptr<ERFilter> createERFilterNM1(const Ptr<ERFilter::Callback>& cb, int threshold
Ptr<ERFilterNM> filter = makePtr<ERFilterNM>();
if (cb == NULL)
filter->setCallback(makePtr<ERClassifierNM1>());
else
filter->setCallback(cb);
filter->setThresholdDelta(thresholdDelta);
@ -1123,7 +1098,8 @@ Ptr<ERFilter> createERFilterNM1(const Ptr<ERFilter::Callback>& cb, int threshold
additional features: hole area ratio, convex hull ratio, and number of outer inflexion points.
\param cb Callback with the classifier
if omitted tries to load a default classifier from file trained_classifierNM2.xml
default classifier can be implicitly load with function loadClassifierNM1()
from file in samples/cpp/trained_classifierNM2.xml
\param minProbability The minimum probability P(er|character) allowed for retreived ER's
*/
Ptr<ERFilter> createERFilterNM2(const Ptr<ERFilter::Callback>& cb, float minProbability)
@ -1133,15 +1109,33 @@ Ptr<ERFilter> createERFilterNM2(const Ptr<ERFilter::Callback>& cb, float minProb
Ptr<ERFilterNM> filter = makePtr<ERFilterNM>();
if (cb == NULL)
filter->setCallback(makePtr<ERClassifierNM2>());
else
filter->setCallback(cb);
filter->setMinProbability(minProbability);
return (Ptr<ERFilter>)filter;
}
/*!
Allow to implicitly load the default classifier when creating an ERFilter object.
The function takes as parameter the XML or YAML file with the classifier model
(e.g. trained_classifierNM1.xml) returns a pointer to ERFilter::Callback.
*/
Ptr<ERFilter::Callback> loadClassifierNM1(const std::string& filename)
{
return makePtr<ERClassifierNM1>(filename);
}
/*!
Allow to implicitly load the default classifier when creating an ERFilter object.
The function takes as parameter the XML or YAML file with the classifier model
(e.g. trained_classifierNM2.xml) returns a pointer to ERFilter::Callback.
*/
Ptr<ERFilter::Callback> loadClassifierNM2(const std::string& filename)
{
return makePtr<ERClassifierNM2>(filename);
}
/* ------------------------------------------------------------------------------------*/
/* -------------------------------- Compute Channels NM -------------------------------*/

4
samples/cpp/erfilter.cpp
View File

@ -58,7 +58,7 @@ int main(int argc, const char * argv[])
double t = (double)getTickCount();
// Build ER tree and filter with the 1st stage default classifier
Ptr<ERFilter> er_filter1 = createERFilterNM1();
Ptr<ERFilter> er_filter1 = createERFilterNM1(loadClassifierNM1("trained_classifierNM1.xml"));
er_filter1->run(grey, regions);
@ -89,7 +89,7 @@ int main(int argc, const char * argv[])
t = (double)getTickCount();
// Default second stage classifier
Ptr<ERFilter> er_filter2 = createERFilterNM2();
Ptr<ERFilter> er_filter2 = createERFilterNM2(loadClassifierNM2("trained_classifierNM2.xml"));
er_filter2->run(grey, regions);
t = (double)getTickCount() - t;