i/rtabmap
1
0
Fork 0
mirror of https://github.com/introlab/rtabmap.git synced 2026-09-03 10:00:23 +08:00
Code Issues Packages Projects Releases Wiki Activity

MarkerDetection: added automatic marker's length estimation when depth image is provided (and when Aruco/MarkerLength is 0).

Browse Source
This commit is contained in:
matlabbe
2019-02-22 20:01:09 -05:00
parent e635f35cf1
commit d089e95e5a
6 changed files with 111 additions and 12 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
corelib/include/rtabmap/core/MarkerDetector.h
View File

@@ -43,7 +43,7 @@ public:
MarkerDetector(const ParametersMap & parameters = ParametersMap());
virtual ~MarkerDetector();
void parseParameters(const ParametersMap & parameters);
std::map<int, Transform> detect(const cv::Mat & image, const CameraModel & model, cv::Mat * imageWithDetections = 0);
std::map<int, Transform> detect(const cv::Mat & image, const CameraModel & model, const cv::Mat & depth = cv::Mat(), float * estimatedMarkerLength = 0, cv::Mat * imageWithDetections = 0);
private:
#ifdef HAVE_OPENCV_ARUCO

2
corelib/include/rtabmap/core/Parameters.h
View File

@@ -713,7 +713,7 @@ class RTABMAP_EXP Parameters
RTABMAP_PARAM(Aruco, Dictionary, int, 0, "Dictionary to use: DICT_4X4_50=0, DICT_4X4_100=1, DICT_4X4_250=2, DICT_4X4_1000=3, DICT_5X5_50=4, DICT_5X5_100=5, DICT_5X5_250=6, DICT_5X5_1000=7, DICT_6X6_50=8, DICT_6X6_100=9, DICT_6X6_250=10, DICT_6X6_1000=11, DICT_7X7_50=12, DICT_7X7_100=13, DICT_7X7_250=14, DICT_7X7_1000=15, DICT_ARUCO_ORIGINAL = 16, DICT_APRILTAG_16h5=17, DICT_APRILTAG_25h9=18, DICT_APRILTAG_36h10=19, DICT_APRILTAG_36h11=20");
RTABMAP_PARAM(Aruco, MarkerLength, float, 0.1, "The length (m) of the markers' side.");
RTABMAP_PARAM(Aruco, VarianceLinear, float, 0.001, "Linear variance to set on marker detections.");
RTABMAP_PARAM(Aruco, VarianceAngular, float, 0.001, "Angular variance to set on marker detections. Set to >=9999 to use only position (xyz) constraint in graph optimization.");
RTABMAP_PARAM(Aruco, VarianceAngular, float, 0.01, "Angular variance to set on marker detections. Set to >=9999 to use only position (xyz) constraint in graph optimization.");
RTABMAP_PARAM(Aruco, CornerRefinementMethod, int, 0, "Corner refinement method (0: None, 1: Subpixel, 2:contour, 3: AprilTag 2). For OpenCV <3.3.0, this is \"doCornerRefinement\" parameter: set 0 for false and 1 for true.");
public:

100
corelib/src/MarkerDetector.cpp
View File

@@ -26,6 +26,7 @@ SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
*/
#include <rtabmap/core/MarkerDetector.h>
#include <rtabmap/core/util2d.h>
#include <rtabmap/utilite/ULogger.h>
namespace rtabmap {
@@ -104,7 +105,7 @@ void MarkerDetector::parseParameters(const ParametersMap & parameters)
#endif
}
std::map<int, Transform> MarkerDetector::detect(const cv::Mat & image, const CameraModel & model, cv::Mat * imageWithDetections)
std::map<int, Transform> MarkerDetector::detect(const cv::Mat & image, const CameraModel & model, const cv::Mat & depth, float * markerLengthOut, cv::Mat * imageWithDetections)
{
std::map<int, Transform> detections;
@@ -123,9 +124,69 @@ std::map<int, Transform> MarkerDetector::detect(const cv::Mat & image, const Cam
UDEBUG("Markers detected=%d rejected=%d", (int)ids.size(), (int)rejected.size());
if(ids.size() > 0)
{
cv::aruco::estimatePoseSingleMarkers(corners, markerLength_, model.K(), model.D(), rvecs, tvecs);
float rgbToDepthFactorX = 1.0f;
float rgbToDepthFactorY = 1.0f;
if(markerLength_ == 0)
{
if(depth.empty())
{
UERROR("Depth image is empty, please set %s parameter to non-null.", Parameters::kArucoMarkerLength().c_str());
return detections;
}
rgbToDepthFactorX = 1.0f/(model.imageWidth()>0?model.imageWidth()/depth.cols:1);
rgbToDepthFactorY = 1.0f/(model.imageHeight()>0?model.imageHeight()/depth.rows:1);
}
cv::aruco::estimatePoseSingleMarkers(corners, markerLength_==0?1.0f:markerLength_, model.K(), model.D(), rvecs, tvecs);
std::vector<float> scales;
for(size_t i=0; i<ids.size(); ++i)
{
if(markerLength_ == 0)
{
float d1 = util2d::getDepth(depth, corners[i][0].x*rgbToDepthFactorX, corners[i][0].y*rgbToDepthFactorY, true, 0.02f, true);
float d2 = util2d::getDepth(depth, corners[i][1].x*rgbToDepthFactorX, corners[i][1].y*rgbToDepthFactorY, true, 0.02f, true);
float d3 = util2d::getDepth(depth, corners[i][2].x*rgbToDepthFactorX, corners[i][2].y*rgbToDepthFactorY, true, 0.02f, true);
float d4 = util2d::getDepth(depth, corners[i][3].x*rgbToDepthFactorX, corners[i][3].y*rgbToDepthFactorY, true, 0.02f, true);
// Accept measurement only if all 4 depth values are valid and
// they are at the same depth (camera should be perpendicular to marker for
// best depth estimation)
if(d1>0 && d2>0 && d3>0 && d4>0)
{
if( fabs(d1-d2) < 0.01f &&
fabs(d1-d3) < 0.01f &&
fabs(d1-d4) < 0.01f)
{
float depth = (d1+d2+d3+d4)/4.0f;
scales.push_back(depth/tvecs[i].val[2]);
tvecs[i] *= scales.back();
UWARN("Automatic marker length estimation: id=%d depth=%fm length=%fm", ids[i], depth, scales.back());
}
else
{
UWARN("The four marker's corners should be "
"perpendicular to camera to estimate correctly "
"the marker's length. Errors: %f, %f, %f > 0.01m."
"Parameter %s can be set to non-null to skip automatic "
"marker length estimation. Detections are ignored.",
fabs(d1-d2), fabs(d1-d3), fabs(d1-d4),
Parameters::kArucoMarkerLength().c_str());
detections.clear();
return detections;
}
}
else
{
UWARN("Some depth values (%f,%f,%f,%f) cannot be detected on the "
"marker's corners, cannot initialize marker length. "
"Parameter %s can be set to non-null to skip automatic "
"marker length estimation. Detections are ignored.",
d1,d2,d3,d4,
Parameters::kArucoMarkerLength().c_str());
detections.clear();
return detections;
}
}
cv::Mat R;
cv::Rodrigues(rvecs[i], R);
Transform t(R.at<double>(0,0), R.at<double>(0,1), R.at<double>(0,2), tvecs[i].val[0],
@@ -136,6 +197,41 @@ std::map<int, Transform> MarkerDetector::detect(const cv::Mat & image, const Cam
detections.insert(std::make_pair(ids[i], pose));
UDEBUG("Marker %d detected at %s (%s)", ids[i], pose.prettyPrint().c_str(), t.prettyPrint().c_str());
}
if(markerLength_ == 0)
{
float sum = 0.0f;
float maxError = 0.0f;
for(size_t i=0; i<scales.size(); ++i)
{
if(i>0)
{
float error = fabs(scales[i]-scales[0]);
if(error > 0.001f)
{
UWARN("The marker's length detected between 2 of the "
"markers doesn't match (%d->%fm vs %d->%fm)."
"Parameter %s can be set to non-null to skip automatic "
"marker length estimation. Detections are ignored.",
ids[i], scales[i], ids[0], scales[0],
Parameters::kArucoMarkerLength().c_str());
detections.clear();
return detections;
}
if(error > maxError)
{
maxError = error;
}
}
sum += scales[i];
}
markerLength_ = sum/float(scales.size());
UWARN("Final marker length estimated = %fm, max error=%fm (used for subsequent detections)", markerLength_, maxError);
}
}
if(markerLengthOut)
{
*markerLengthOut = markerLength_;
}
if(imageWithDetections)

7
corelib/src/Memory.cpp
View File

@@ -4604,11 +4604,14 @@ Signature * Memory::createSignature(const SensorData & inputData, const Transfor
static bool warned = false;
if(!warned)
{
UWARN("Detecting markers in multi-camera setup is not yet implemented, detecting only in first camera. This message is only printed once.");
UWARN("Detecting markers in multi-camera setup is not yet implemented, aborting marker detection. This message is only printed once.");
}
warned = true;
}
markers = _markerDetector->detect(data.imageRaw(), data.cameraModels()[0]);
else
{
markers = _markerDetector->detect(data.imageRaw(), data.cameraModels()[0], data.depthRaw());
}
}
else if(data.stereoCameraModel().isValidForProjection())
{