VDT.FaceRecognition.SDK.CameraCalibrator Class Reference

Interface object for camera calibration and correction of image distortion. More...

Classes
class	AddImageOutput
	Result of image processing. More...
class	CalibrateOutput
	Result of calibration. More...
class	CalibrationSettings
	Сalibration process settings. More...

Public Member Functions
void	initCalibration (CalibrationSettings settings)
	Initialize the calibration process. All previous accepted patterns will be removed. More...
AddImageOutput	addImage (RawImage image)
	Search a pattern in the image and store it. More...
float	getPatternSpaceCoverProgress ()
	Evaluate covering of the pattern space to get more accurate calibration. More...
List< Point >	getTip ()
	Get a tip where to put a pattern in the image to get a better pattern space coverage. More...
CalibrateOutput	calibrate (int max_used_patterns_count)
	Calibrate the camera. More...
RawImage	undistort (RawImage image, float alpha)
	Correct the image distortion using the calibrated or loaded camera parameters. More...
void	saveCameraParameters (Stream binary_stream)
	Save the calibrated camera parameters to binary_stream. The format is platform-independent. More...
void	loadCameraParameters (Stream binary_stream)
	Load the calibrated camera parameters from binary_stream. The format is platform-independent. More...
Public Member Functions inherited from VDT.FaceRecognition.SDK.IDisposableObject
virtual void	Dispose ()
	Release internal object.

Additional Inherited Members
Properties inherited from VDT.FaceRecognition.SDK.IDisposableObject
bool	isDisposed [get]
	True if internal object already disposed.

Detailed Description

Interface object for camera calibration and correction of image distortion.

Member Function Documentation

AddImageOutput VDT.FaceRecognition.SDK.CameraCalibrator.addImage ( RawImage  image )

inline

Search a pattern in the image and store it.

Parameters

[in]

image

An image or a videoframe.

Returns

Result of image processing.

CalibrateOutput VDT.FaceRecognition.SDK.CameraCalibrator.calibrate ( int  max_used_patterns_count )

inline

Calibrate the camera.

Parameters

[in]

max_used_patterns_count

Maximum number of patterns to use for calculation. If more patterns are accepted, the subset covering the pattern space best will be used. The recommended value is 50.

Returns

Result of calibration.

float VDT.FaceRecognition.SDK.CameraCalibrator.getPatternSpaceCoverProgress ( )

inline

Evaluate covering of the pattern space to get more accurate calibration.

Returns

A real number from 0 to 1.

List<Point> VDT.FaceRecognition.SDK.CameraCalibrator.getTip ( )

inline

Get a tip where to put a pattern in the image to get a better pattern space coverage.

Returns

An empty list if the tip is not ready, or a list of pattern points.

void VDT.FaceRecognition.SDK.CameraCalibrator.initCalibration ( CalibrationSettings  settings )

inline

Initialize the calibration process. All previous accepted patterns will be removed.

Parameters

[in]

settings

Calibration process settings.

void VDT.FaceRecognition.SDK.CameraCalibrator.loadCameraParameters ( Stream  binary_stream )

inline

Load the calibrated camera parameters from binary_stream. The format is platform-independent.

Parameters

[in]

binary_stream

Input stream object.

void VDT.FaceRecognition.SDK.CameraCalibrator.saveCameraParameters ( Stream  binary_stream )

inline

Save the calibrated camera parameters to binary_stream. The format is platform-independent.

Parameters

[out]

binary_stream

Output stream object.

RawImage VDT.FaceRecognition.SDK.CameraCalibrator.undistort ( RawImage  image, float  alpha  )

inline

Correct the image distortion using the calibrated or loaded camera parameters.

Parameters

[in]	image	An image or a videoframe. The size can differ from the size of the images used for calibration. I.e. you can calibrate a camera once with one resolution and then calibrate the distortion of images from the same camera with any resolution.
[in]	alpha	A real value from 0 (the result image contains only valid pixels (no black areas)) to 1 (all pixels from the original image are stored in the result image (no source image pixels are lost)). Intermediate values yield an intermediate result between those two extreme cases.

Returns

Undistorted image of the same size and type as the source image.