org.openimaj.hardware.kinect

Class KinectController

java.lang.Object

org.openimaj.hardware.kinect.KinectController

public class KinectController
extends Object

The OpenIMAJ Kinect Interface.

Author:

Jonathon Hare (jsh2@ecs.soton.ac.uk)

Field Summary

Fields

Modifier and Type	Field and Description
protected static List<KinectController>	ACTIVE_CONTROLLERS
protected static org.bridj.Pointer<libfreenectLibrary.freenect_context>	CONTEXT
KinectDepthStream	depthStream The depth stream
protected org.bridj.Pointer<libfreenectLibrary.freenect_device>	device
protected static org.openimaj.hardware.kinect.EventThread	EVENT_THREAD
KinectStream<?>	videoStream The RGB or IR video stream

Constructor Summary

Constructors

Constructor and Description
KinectController() Default constructor.
KinectController(boolean irmode) Construct with the first device in the given mode.
KinectController(int deviceId) Construct with given device
KinectController(int deviceId, boolean irmode, boolean registeredDepthMode) Construct with the given device and mode.

Method Summary

Modifier and Type	Method and Description
float[][]	cameraToWorld(FImage regDepth, int startx, int stopx, int stepx, int starty, int stopy, int stepy) Compute the world coordinates from the pixel locations in the given registered depth image.
double[]	cameraToWorld(int x, int y, int depth) Compute the world coordinates from the pixel location and registered depth.
float[]	cameraToWorld(int x, int y, int depth, double factor) Compute the world coordinates from the pixel location and registered depth.
float[]	cameraToWorld(int x, int y, int depth, double factor, float[] pt) Compute the world coordinates from the pixel location and registered depth.
void	close() Close the device.
double	computeScalingFactor() Compute the scaling factor for computing world coordinates.
static int	connectedDevices() Get the number of connected devices.
void	finalize()
KinectAcceleration	getAcceleration() Get the current acceleration values from the device
double	getTilt() Get the current angle in degrees
KinectTiltStatus	getTiltStatus() Determine the current tilt status of the device
void	identify() Sets the LEDs to blink red and yellow for five seconds before resetting to green.
static void	main(String[] args) Test
void	setIRMode(boolean irmode) Switch the main camera between InfraRed and RGB modes.
void	setLED(KinectLEDMode option) Set the device status LEDs
void	setRegisteredDepth(boolean rdepth) Set whether depth should be registered
void	setTilt(double angle) Set the tilt to the given angle in degrees
static void	shutdownFreenect() Completely shutdown the context.

Methods inherited from class java.lang.Object

clone, equals, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Field Detail

CONTEXT

protected static volatile org.bridj.Pointer<libfreenectLibrary.freenect_context> CONTEXT

EVENT_THREAD

protected static volatile org.openimaj.hardware.kinect.EventThread EVENT_THREAD

ACTIVE_CONTROLLERS

protected static volatile List<KinectController> ACTIVE_CONTROLLERS

device

protected org.bridj.Pointer<libfreenectLibrary.freenect_device> device

videoStream

public KinectStream<?> videoStream

The RGB or IR video stream

depthStream

public KinectDepthStream depthStream

The depth stream

Constructor Detail

KinectController

public KinectController(int deviceId)
                 throws KinectException

Construct with given device

Parameters:

deviceId - the device id

Throws:

KinectException

KinectController

public KinectController(boolean irmode)
                 throws KinectException

Construct with the first device in the given mode.

Parameters:

irmode - if true then the camera is set to IR mode; otherwise its in RGB mode

Throws:

KinectException

KinectController

public KinectController()
                 throws KinectException

Default constructor. Uses the first device in RGB mode.

Throws:

KinectException

KinectController

public KinectController(int deviceId,
                        boolean irmode,
                        boolean registeredDepthMode)
                 throws KinectException

Construct with the given device and mode.

Parameters:

deviceId - the device identifier. 0 for the first one.

irmode - whether to use infra-red mode or rgb mode.

registeredDepthMode - whether to register the depth image. If true, depth measurements are in millimeters.

Throws:

KinectException

Method Detail

finalize

public void finalize()

Overrides:

finalize in class Object

shutdownFreenect

public static void shutdownFreenect()

Completely shutdown the context. This turns off all cameras. The context will be restarted upon creation of a new KinectController.

setIRMode

public void setIRMode(boolean irmode)

Switch the main camera between InfraRed and RGB modes.

Parameters:

irmode - if true, then switches to IR mode, otherwise switches to RGB mode.

setRegisteredDepth

public void setRegisteredDepth(boolean rdepth)

Set whether depth should be registered

Parameters:

rdepth - if true, then switches to depth registered mode, otherwise depth is not registered

connectedDevices

public static int connectedDevices()
                            throws KinectException

Get the number of connected devices.

Returns:

the number of devices connected.

Throws:

KinectException

close

public void close()

Close the device. Closing an already closed device has no effect.

getTilt

public double getTilt()

Get the current angle in degrees

Returns:

the angle or 0 if the device is closed

setTilt

public void setTilt(double angle)

Set the tilt to the given angle in degrees

Parameters:

angle - the angle

getTiltStatus

public KinectTiltStatus getTiltStatus()

Determine the current tilt status of the device

Returns:

the tilt status or null if the device is closed

setLED

public void setLED(KinectLEDMode option)

Set the device status LEDs

Parameters:

option - the LED status to set

identify

public void identify()

Sets the LEDs to blink red and yellow for five seconds before resetting to green.

getAcceleration

public KinectAcceleration getAcceleration()

Get the current acceleration values from the device

Returns:

the acceleration or null if the device is closed

cameraToWorld

public double[] cameraToWorld(int x,
                              int y,
                              int depth)

Compute the world coordinates from the pixel location and registered depth.

Parameters:

x - pixel x-ordinate

y - pixel y-ordinate

depth - the depth

Returns:

the (x,y,z) coordinate in world space

computeScalingFactor

public double computeScalingFactor()

Compute the scaling factor for computing world coordinates.

Returns:

the scaling factor

See Also:

cameraToWorld(int, int, int, double)

cameraToWorld

public float[] cameraToWorld(int x,
                             int y,
                             int depth,
                             double factor)

Compute the world coordinates from the pixel location and registered depth. This method requires you pre-compute the constant scaling factor using computeScalingFactor(), but it should be faster than using cameraToWorld(int, int, int).

Parameters:

x - pixel x-ordinate

y - pixel y-ordinate

depth - the depth

factor - the scaling factor

Returns:

the (x,y,z) coordinate in world space

cameraToWorld

public float[] cameraToWorld(int x,
                             int y,
                             int depth,
                             double factor,
                             float[] pt)

Compute the world coordinates from the pixel location and registered depth. This method requires you pre-compute the constant scaling factor using computeScalingFactor(), but it should be faster than using cameraToWorld(int, int, int). This method is the same as cameraToWorld(int, int, int, double), but reuses the point array for efficiency.

Parameters:

x - pixel x-ordinate

y - pixel y-ordinate

depth - the depth

factor -

pt - the point to write to

Returns:

the (x,y,z) coordinate in world space

cameraToWorld

public float[][] cameraToWorld(FImage regDepth,
                               int startx,
                               int stopx,
                               int stepx,
                               int starty,
                               int stopy,
                               int stepy)

Compute the world coordinates from the pixel locations in the given registered depth image. This method basically gives you a fully registered point cloud.

Parameters:

regDepth - the registered depth image

startx - the starting x-ordinate in the image

stopx - the stopping x-ordinate in the image

stepx - the number of pixels in the x direction to skip between samples

starty - the starting y-ordinate in the image

stopy - the stopping y-ordinate in the image

stepy - the number of pixels in the y direction to skip between samples

Returns:

an array of 3d vectors

main

public static void main(String[] args)
                 throws KinectException

Test

Parameters:

args -

Throws:

KinectException