boofcv.abst.fiducial

Interface FiducialDetector<T extends boofcv.struct.image.ImageBase<T>>

All Known Implementing Classes:

CalibrationFiducialDetector, FiducialDetectorPnP, SquareBase_to_FiducialDetector, SquareBinary_to_FiducialDetector, SquareImage_to_FiducialDetector

public interface FiducialDetector<T extends boofcv.struct.image.ImageBase<T>>

Interface for detecting fiducial markers and their location in the image. Optionally, some implementations can produce a unique ID for each fiducial and their 3D pose in the world. If the fiducial supports 3D information then

Method Summary

Modifier and Type	Method and Description
boolean	computeStability(int which, double disturbance, FiducialStability results) Computes metrics which represents the fiducial's pose estimate stability given its current observed state.
void	detect(T input) Detects fiducials inside the image
georegression.struct.shapes.Polygon2D_F64	getBounds(int which, georegression.struct.shapes.Polygon2D_F64 storage) Used to retrieve the bounds around a marker in the image.
void	getCenter(int which, georegression.struct.point.Point2D_F64 location) Returns where in the image the fiducial is.
boolean	getFiducialToCamera(int which, georegression.struct.se.Se3_F64 fiducialToCamera) Used to retrieve the transformation from the fiducial's reference frame to the camera's reference frame.
long	getId(int which) If applicable, returns the ID of the fiducial found.
boofcv.struct.image.ImageType<T>	getInputType() Type of input image
boofcv.alg.distort.LensDistortionNarrowFOV	getLensDistortion() Returns the intrinsic parameters that it
java.lang.String	getMessage(int which) If applicable, returns a message associated with the fiducial.
double	getWidth(int which) Returns the width of the fiducial in world units.
boolean	hasMessage() If true then getMessage(int) returns a valid message
boolean	hasUniqueID() If true then getId(int) returns a valid unique number
boolean	is3D() If true then 3D information is available for the fiducial.
void	setLensDistortion(boofcv.alg.distort.LensDistortionNarrowFOV distortion, int width, int height) Specifies how to remove lens distortion from the input image and how to convert pixels into normalized image coordinates.
int	totalFound() The total number of fiducial found

Method Detail

detect

void detect(T input)

Detects fiducials inside the image

Parameters:

input - Input image. Not modified.

totalFound

int totalFound()

The total number of fiducial found

Returns:

number of targets found

getCenter

void getCenter(int which,
               georegression.struct.point.Point2D_F64 location)

Returns where in the image the fiducial is. Typically this will be the fiducial's visual center. Note that the visual center is unlikely to be the projection of the 3D geometric center. To find the former you need to reproject it using the found fiducialToCamera.

NOTE: The reprojected center might not be the same as the location returned here.

Parameters:

which - Fiducial's index

location - (output) Storage for the transform. modified.

getBounds

georegression.struct.shapes.Polygon2D_F64 getBounds(int which,
                                                    @Nullable
                                                    georegression.struct.shapes.Polygon2D_F64 storage)

Used to retrieve the bounds around a marker in the image. How the bounds are defined is left up the implementation. It could be a simple rectangle or it could be corner features.

Parameters:

which - Which fiducial.

storage - (Optional) Storage for markers.

Returns:

Found marker. Points are owned by the caller and will not be modified in the future.

getId

long getId(int which)

If applicable, returns the ID of the fiducial found. Call hasUniqueID() to see if this function returns a valid value.

Parameters:

which - Detected fiducial's index

Returns:

ID of the fiducial

getMessage

java.lang.String getMessage(int which)

If applicable, returns a message associated with the fiducial. Call hasMessage() ()} to see if this function returns a valid value.

Parameters:

which - Detected fiducial's index

Returns:

Encoded message

setLensDistortion

void setLensDistortion(@Nullable
                       boofcv.alg.distort.LensDistortionNarrowFOV distortion,
                       int width,
                       int height)

Specifies how to remove lens distortion from the input image and how to convert pixels into normalized image coordinates.

Parameters:

distortion - Lens distortion model. null if you want to remove a lens distortion model that had previously been set.

width - Input image's width.

height - Input image's height

getLensDistortion

boofcv.alg.distort.LensDistortionNarrowFOV getLensDistortion()

Returns the intrinsic parameters that it

Returns:

intrinsic parameters

computeStability

boolean computeStability(int which,
                         double disturbance,
                         FiducialStability results)

Computes metrics which represents the fiducial's pose estimate stability given its current observed state. This can be viewed as an estimate of the pose estimate's precision, but not its accuracy. These numbers are generated by perturbing landmarks (by the user provided amount) and seeing how it affects the pose estimate..

The metrics should be considered more qualitative than quantitative since exactly how this metric is computed isn't specified and can vary depending on target type of implementation type. The results could every vary each time its called, even with the the exact same inputs.

Parameters:

which - Index of which fiducial the stability is being requested from

disturbance - Amount of the applied disturbance, in pixels. Try 0.25

results - (output) Storage for stability metrics.

Returns:

true if successful or false if it failed for some reason

getFiducialToCamera

boolean getFiducialToCamera(int which,
                            georegression.struct.se.Se3_F64 fiducialToCamera)

Used to retrieve the transformation from the fiducial's reference frame to the camera's reference frame.

Parameters:

which - Fiducial's index

fiducialToCamera - (output) Storage for the transform. modified.

Returns:

true if could estimate the location or false if it couldn't

getWidth

double getWidth(int which)

Returns the width of the fiducial in world units. If not square then it returns a reasonable approximation. Intended for use in visualization and not precise calculations.

Parameters:

which - Fiducial's index

Returns:

Fiducial's width.

is3D

boolean is3D()

If true then 3D information is available for the fiducial. In general a len distortion model must be provided by invoking setLensDistortion(LensDistortionNarrowFOV,int,int). The following functions are then enabled:

1. computeStability(int, double, boofcv.abst.fiducial.FiducialStability)
2. getFiducialToCamera(int, georegression.struct.se.Se3_F64)

Returns:

hasUniqueID

boolean hasUniqueID()

If true then getId(int) returns a valid unique number

Returns:

boolean

hasMessage

boolean hasMessage()

If true then getMessage(int) returns a valid message

Returns:

boolean

getInputType

boofcv.struct.image.ImageType<T> getInputType()

Type of input image