opencv/samples/cpp/detector_descriptor_matcher_evaluation.cpp

347 lines
12 KiB
C++
Raw Normal View History

/*M///////////////////////////////////////////////////////////////////////////////////////
//
// IMPORTANT: READ BEFORE DOWNLOADING, COPYING, INSTALLING OR USING.
//
// By downloading, copying, installing or using the software you agree to this license.
// If you do not agree to this license, do not download, install,
// copy or use the software.
//
//
// Intel License Agreement
// For Open Source Computer Vision Library
//
// Copyright (C) 2000, Intel Corporation, all rights reserved.
// Third party copyrights are property of their respective owners.
//
// Redistribution and use in source and binary forms, with or without modification,
// are permitted provided that the following conditions are met:
//
// * Redistribution's of source code must retain the above copyright notice,
// this list of conditions and the following disclaimer.
//
// * Redistribution's in binary form must reproduce the above copyright notice,
// this list of conditions and the following disclaimer in the documentation
// and/or other materials provided with the distribution.
//
// * The name of Intel Corporation may not be used to endorse or promote products
// derived from this software without specific prior written permission.
//
// This software is provided by the copyright holders and contributors "as is" and
// any express or implied warranties, including, but not limited to, the implied
// warranties of merchantability and fitness for a particular purpose are disclaimed.
// In no event shall the Intel Corporation or contributors be liable for any direct,
// indirect, incidental, special, exemplary, or consequential damages
// (including, but not limited to, procurement of substitute goods or services;
// loss of use, data, or profits; or business interruption) however caused
// and on any theory of liability, whether in contract, strict liability,
// or tort (including negligence or otherwise) arising in any way out of
// the use of this software, even if advised of the possibility of such damage.
//
//M*/
#include "opencv2/core/utility.hpp"
#include "opencv2/imgproc.hpp"
#include "opencv2/highgui.hpp"
#include "opencv2/features2d.hpp"
#include <stdlib.h>
#include <stdio.h>
#include <sys/stat.h>
#include <limits>
#include <cstdio>
#include <iostream>
#include <fstream>
using namespace std;
using namespace cv;
/*
The algorithm:
for each tested combination of detector+descriptor+matcher:
2012-10-17 09:12:04 +02:00
create detector, descriptor and matcher,
load their params if they are there, otherwise use the default ones and save them
2012-10-17 09:12:04 +02:00
for each dataset:
2012-10-17 09:12:04 +02:00
load reference image
detect keypoints in it, compute descriptors
2012-10-17 09:12:04 +02:00
for each transformed image:
load the image
load the transformation matrix
detect keypoints in it too, compute descriptors
2012-10-17 09:12:04 +02:00
find matches
transform keypoints from the first image using the ground-truth matrix
2012-10-17 09:12:04 +02:00
compute the number of matched keypoints, i.e. for each pair (i,j) found by a matcher compare
j-th keypoint from the second image with the transformed i-th keypoint. If they are close, +1.
2012-10-17 09:12:04 +02:00
so, we have:
N - number of keypoints in the first image that are also visible
(after transformation) on the second image
2012-10-17 09:12:04 +02:00
N1 - number of keypoints in the first image that have been matched.
2012-10-17 09:12:04 +02:00
n - number of the correct matches found by the matcher
2012-10-17 09:12:04 +02:00
n/N1 - precision
n/N - recall (?)
2012-10-17 09:12:04 +02:00
we store (N, n/N1, n/N) (where N is stored primarily for tuning the detector's thresholds,
in order to semi-equalize their keypoints counts)
2012-10-17 09:12:04 +02:00
*/
typedef Vec3f TVec; // (N, n/N1, n/N) - see above
static void saveloadDDM( const string& params_filename,
Ptr<FeatureDetector>& detector,
Ptr<DescriptorExtractor>& descriptor,
Ptr<DescriptorMatcher>& matcher )
{
FileStorage fs(params_filename, FileStorage::READ);
if( fs.isOpened() )
{
detector->read(fs["detector"]);
descriptor->read(fs["descriptor"]);
matcher->read(fs["matcher"]);
}
else
{
fs.open(params_filename, FileStorage::WRITE);
fs << "detector" << "{";
detector->write(fs);
fs << "}" << "descriptor" << "{";
descriptor->write(fs);
fs << "}" << "matcher" << "{";
matcher->write(fs);
fs << "}";
}
}
static Mat loadMat(const string& fsname)
{
FileStorage fs(fsname, FileStorage::READ);
Mat m;
fs.getFirstTopLevelNode() >> m;
return m;
}
static void transformKeypoints( const vector<KeyPoint>& kp,
vector<vector<Point2f> >& contours,
const Mat& H )
{
const float scale = 256.f;
size_t i, n = kp.size();
contours.resize(n);
vector<Point> temp;
2012-10-17 09:12:04 +02:00
for( i = 0; i < n; i++ )
{
ellipse2Poly(Point2f(kp[i].pt.x*scale, kp[i].pt.y*scale),
Size2f(kp[i].size*scale, kp[i].size*scale),
0, 0, 360, 12, temp);
Mat(temp).convertTo(contours[i], CV_32F, 1./scale);
perspectiveTransform(contours[i], contours[i], H);
}
}
static TVec proccessMatches( Size imgsize,
const vector<DMatch>& matches,
const vector<vector<Point2f> >& kp1t_contours,
const vector<vector<Point2f> >& kp_contours,
double overlapThreshold )
{
const double visibilityThreshold = 0.6;
2012-10-17 09:12:04 +02:00
// 1. [preprocessing] find bounding rect for each element of kp1t_contours and kp_contours.
// 2. [cross-check] for each DMatch (iK, i1)
// update best_match[i1] using DMatch::distance.
// 3. [compute overlapping] for each i1 (keypoint from the first image) do:
// if i1-th keypoint is outside of image, skip it
// increment N
// if best_match[i1] is initialized, increment N1
// if kp_contours[best_match[i1]] and kp1t_contours[i1] overlap by overlapThreshold*100%,
// increment n. Use bounding rects to speedup this step
2012-10-17 09:12:04 +02:00
int i, size1 = (int)kp1t_contours.size(), size = (int)kp_contours.size(), msize = (int)matches.size();
vector<DMatch> best_match(size1);
vector<Rect> rects1(size1), rects(size);
2012-10-17 09:12:04 +02:00
// proprocess
for( i = 0; i < size1; i++ )
rects1[i] = boundingRect(kp1t_contours[i]);
2012-10-17 09:12:04 +02:00
for( i = 0; i < size; i++ )
rects[i] = boundingRect(kp_contours[i]);
2012-10-17 09:12:04 +02:00
// cross-check
for( i = 0; i < msize; i++ )
{
DMatch m = matches[i];
int i1 = m.trainIdx, iK = m.queryIdx;
CV_Assert( 0 <= i1 && i1 < size1 && 0 <= iK && iK < size );
if( best_match[i1].trainIdx < 0 || best_match[i1].distance > m.distance )
best_match[i1] = m;
}
2012-10-17 09:12:04 +02:00
int N = 0, N1 = 0, n = 0;
2012-10-17 09:12:04 +02:00
// overlapping
for( i = 0; i < size1; i++ )
{
int i1 = i, iK = best_match[i].queryIdx;
if( iK >= 0 )
N1++;
2012-10-17 09:12:04 +02:00
Rect r = rects1[i] & Rect(0, 0, imgsize.width, imgsize.height);
if( r.area() < visibilityThreshold*rects1[i].area() )
continue;
N++;
2012-10-17 09:12:04 +02:00
if( iK < 0 || (rects1[i1] & rects[iK]).area() == 0 )
continue;
2012-10-17 09:12:04 +02:00
double n_area = intersectConvexConvex(kp1t_contours[i1], kp_contours[iK], noArray(), true);
if( n_area == 0 )
continue;
2012-10-17 09:12:04 +02:00
double area1 = contourArea(kp1t_contours[i1], false);
double area = contourArea(kp_contours[iK], false);
2012-10-17 09:12:04 +02:00
double ratio = n_area/(area1 + area - n_area);
n += ratio >= overlapThreshold;
}
2012-10-17 09:12:04 +02:00
return TVec((float)N, (float)n/std::max(N1, 1), (float)n/std::max(N, 1));
}
static void saveResults(const string& dir, const string& name, const string& dsname,
const vector<TVec>& results, const int* xvals)
{
string fname1 = format("%s%s_%s_precision.csv", dir.c_str(), name.c_str(), dsname.c_str());
string fname2 = format("%s%s_%s_recall.csv", dir.c_str(), name.c_str(), dsname.c_str());
FILE* f1 = fopen(fname1.c_str(), "wt");
FILE* f2 = fopen(fname2.c_str(), "wt");
2012-10-17 09:12:04 +02:00
for( size_t i = 0; i < results.size(); i++ )
{
fprintf(f1, "%d, %.1f\n", xvals[i], results[i][1]*100);
fprintf(f2, "%d, %.1f\n", xvals[i], results[i][2]*100);
}
fclose(f1);
fclose(f2);
}
int main(int argc, char** argv)
{
static const char* ddms[] =
{
"ORBX_BF", "ORB", "ORB", "BruteForce-Hamming",
//"ORB_BF", "ORB", "ORB", "BruteForce-Hamming",
//"ORB3_BF", "ORB", "ORB", "BruteForce-Hamming(2)",
//"ORB4_BF", "ORB", "ORB", "BruteForce-Hamming(2)",
//"ORB_LSH", "ORB", "ORB", "LSH"
//"SURF_BF", "SURF", "SURF", "BruteForce",
0
};
2012-10-17 09:12:04 +02:00
static const char* datasets[] =
{
"bark", "bikes", "boat", "graf", "leuven", "trees", "ubc", "wall", 0
};
2012-10-17 09:12:04 +02:00
static const int imgXVals[] = { 2, 3, 4, 5, 6 }; // if scale, blur or light changes
static const int viewpointXVals[] = { 20, 30, 40, 50, 60 }; // if viewpoint changes
static const int jpegXVals[] = { 60, 80, 90, 95, 98 }; // if jpeg compression
2012-10-17 09:12:04 +02:00
const double overlapThreshold = 0.6;
2012-10-17 09:12:04 +02:00
vector<vector<vector<TVec> > > results; // indexed as results[ddm][dataset][testcase]
2012-10-17 09:12:04 +02:00
string dataset_dir = string(getenv("OPENCV_TEST_DATA_PATH")) +
"/cv/detectors_descriptors_evaluation/images_datasets";
2012-10-17 09:12:04 +02:00
string dir=argc > 1 ? argv[1] : ".";
2012-10-17 09:12:04 +02:00
if( dir[dir.size()-1] != '\\' && dir[dir.size()-1] != '/' )
dir += "/";
2012-10-17 09:12:04 +02:00
int result = system(("mkdir " + dir).c_str());
CV_Assert(result == 0);
2012-10-17 09:12:04 +02:00
for( int i = 0; ddms[i*4] != 0; i++ )
{
const char* name = ddms[i*4];
const char* detector_name = ddms[i*4+1];
const char* descriptor_name = ddms[i*4+2];
const char* matcher_name = ddms[i*4+3];
string params_filename = dir + string(name) + "_params.yml";
2012-10-17 09:12:04 +02:00
cout << "Testing " << name << endl;
2012-10-17 09:12:04 +02:00
Ptr<FeatureDetector> detector = FeatureDetector::create(detector_name);
Ptr<DescriptorExtractor> descriptor = DescriptorExtractor::create(descriptor_name);
Ptr<DescriptorMatcher> matcher = DescriptorMatcher::create(matcher_name);
2012-10-17 09:12:04 +02:00
saveloadDDM( params_filename, detector, descriptor, matcher );
2012-10-17 09:12:04 +02:00
results.push_back(vector<vector<TVec> >());
2012-10-17 09:12:04 +02:00
for( int j = 0; datasets[j] != 0; j++ )
{
const char* dsname = datasets[j];
2012-10-17 09:12:04 +02:00
cout << "\ton " << dsname << " ";
cout.flush();
2012-10-17 09:12:04 +02:00
const int* xvals = strcmp(dsname, "ubc") == 0 ? jpegXVals :
strcmp(dsname, "graf") == 0 || strcmp(dsname, "wall") == 0 ? viewpointXVals : imgXVals;
2012-10-17 09:12:04 +02:00
vector<KeyPoint> kp1, kp;
vector<DMatch> matches;
vector<vector<Point2f> > kp1t_contours, kp_contours;
Mat desc1, desc;
2012-10-17 09:12:04 +02:00
Mat img1 = imread(format("%s/%s/img1.png", dataset_dir.c_str(), dsname), 0);
CV_Assert( !img1.empty() );
2012-10-17 09:12:04 +02:00
detector->detect(img1, kp1);
descriptor->compute(img1, kp1, desc1);
2012-10-17 09:12:04 +02:00
results[i].push_back(vector<TVec>());
2012-10-17 09:12:04 +02:00
for( int k = 2; ; k++ )
{
cout << ".";
cout.flush();
Mat imgK = imread(format("%s/%s/img%d.png", dataset_dir.c_str(), dsname, k), 0);
if( imgK.empty() )
break;
2012-10-17 09:12:04 +02:00
detector->detect(imgK, kp);
descriptor->compute(imgK, kp, desc);
matcher->match( desc, desc1, matches );
2012-10-17 09:12:04 +02:00
Mat H = loadMat(format("%s/%s/H1to%dp.xml", dataset_dir.c_str(), dsname, k));
2012-10-17 09:12:04 +02:00
transformKeypoints( kp1, kp1t_contours, H );
transformKeypoints( kp, kp_contours, Mat::eye(3, 3, CV_64F));
2012-10-17 09:12:04 +02:00
TVec r = proccessMatches( imgK.size(), matches, kp1t_contours, kp_contours, overlapThreshold );
results[i][j].push_back(r);
}
2012-10-17 09:12:04 +02:00
saveResults(dir, name, dsname, results[i][j], xvals);
cout << endl;
}
}
}