VisualStimulus Class Reference

Class to integrate CARLsim with a stimulus created using VisualStimulus.m Version: 4/11/14 Author: Michael Beyeler mbeye.nosp@m.ler@.nosp@m.uci.e.nosp@m.du More...

#include <visual_stimulus.h>

Public Member Functions
	VisualStimulus (std::string fileName, bool wrapAroundEOF=true)
	Default constructor. More...
	~VisualStimulus ()
	default destructor More...
int	getChannels ()
	returns the number of channels (1=grayscale, 3=RGB) More...
unsigned char *	getCurrentFrameChar ()
	returns char array of current frame More...
int	getCurrentFrameNumber ()
PoissonRate *	getCurrentFramePoisson ()
	returns PoissonRate object of current frame More...
int	getHeight ()
	returns the stimulus height (number of pixels) More...
int	getLength ()
	returns the stimulus length (number of frames) More...
stimType_t	getType ()
	returns the stimulus type (STIM_GRAY, STIM_RGB, etc.) More...
int	getWidth ()
	returns the stimulus width (number of pixels) More...
void	print ()
unsigned char *	readFrameChar ()
	Reads the next image frame and returns a pointer to the char array. More...
PoissonRate *	readFramePoisson (float maxPoisson, float minPoisson=0.0f)
	Reads the next image frame and returns a pointer to a PoissonRate object. More...
void	rewind ()
	Rewinds the file pointer to the top. More...

Detailed Description

This class provides a means to integrate CARLsim with a stimulus created using VisualStimulus.m. The class reads from binary file, and returns either the raw character array or a PoissonRate object with the same size as number of pixels in a stimulus frame.

Common workflow:

• 1. Create a stimulus in Matlab:

  * >> VS = VisualStimulus(32,32);
  * >> VS.addSinGrating
  * >> VS.saveToFile
  * 

  Successfully saved stimulus to file "inpGrating_gray_32x32x80.dat"

• 2. In your main_.cpp, read from binary file:

  VisualStimulus VS("inpGrating_gray_32x32x80.dat");
  int videoLength = VS.getStimulusLength();

• 3. Assign each frame to a SpikeGenerator group and run network

  for (int i=0; i<videoLength; i++) {
      PoissonRate * rates = VS.readFrame(50.0f); // grayscale value 255 will be mapped to 50 Hz
      snn.setSpikeRate(g1, rates); // for this to work, there must be 32x32=1024 neurons in g1
      snn.runNetwork(1,0); // run the network
  }

Definition at line 56 of file visual_stimulus.h.

Constructor & Destructor Documentation

VisualStimulus	(	std::string	fileName,
		bool	wrapAroundEOF = true
	)

Instantiates an VisualStimulus object.

Parameters

[in]	fileName	path to binary file that was created using VisualStimulus.m
[in]	wrapAroundEOF	after reaching the end of the file, whether to jump back to the top and start reading the first frame again. If this flag is false and the EOF is reached, a warning will be printed. Default: true.

~VisualStimulus

(

)

Member Function Documentation

int getChannels

(

)

unsigned char* getCurrentFrameChar

(

)

int getCurrentFrameNumber

(

)

PoissonRate* getCurrentFramePoisson

(

)

int getHeight

(

)

int getLength

(

)

stimType_t getType

(

)

int getWidth

(

)

void print

(

)

unsigned char* readFrameChar

(

)

Advances the frame index by 1 (getCurrentFrameNumber) and returns the raw grayscale values for each pixel. The char array will have a total of getStimulusWidth()*getStimulusHeight()*getStimulusChannels() entries. The order is columns first (so the first getStimulusWidth() number of pixels will correspond to the top image row). Currently, only grayscale images are supported. Note that this will advance the frame index. If you want to access the char array or PoissonRate object of a frame that has already been read, use getCurrentFrameChar() or getCurrentFramePoisson() instead.

Returns

pointer to the char array of raw grayscale values

PoissonRate* readFramePoisson	(	float	maxPoisson,
		float	minPoisson = 0.0f
	)

Advances the frame index by 1 (getCurrentFrameNumber) and returns a pointer to a PoissonRate object. The object will have a total of getStimulusWidth()*getStimulusHeight()*getStimulusChannels() entries. The order is columns first (so the first getStimulusWidth() number of pixels will correspond to the top image row). The PoissonRate object can then be applied to a group: PoissonRate* rates = IS.readFrame(50.0f); snn.setSpikeRate(g1, rates); Currently, only grayscale images are supported.

Parameters

[in]	maxPoisson	Maximum Poisson rate (must be positive). The range of grayscale values [0,255] will be linearly mapped to the range of Poisson rates [minPoisson,maxPoisson].
[in]	minPoisson	Maximum Poisson rate (must be non-negative). The range of grayscale values [0,255] will be linearly mapped to the range of Poisson rates [minPoisson,maxPoisson]. Default: 0 Hz.

Returns

pointer to a PoissonRate object

Note

maxPoisson must be greater than minPoisson. Neither of them can be

Attention

Each call to readFrame() will advance the frame index. If you want to access the char array or PoissonRate object of a frame that has already been read, use getCurrentFrameChar() or getCurrentFramePoisson() instead.

void rewind

(

)

This function rewinds the file pointer back to the beginning of the file, so that the user can re-start reading the stimulus from the top.

---

The documentation for this class was generated from the following file:

• visual_stimulus.h