PerlinNoise Class Reference

Class used to generate 2D perlin noise. More...

#include <PerlinNoise.h>

List of all members.

Public Member Functions
	PerlinNoise (int size_x, int size_y)
	~PerlinNoise ()
void	SetForegroundColor (float red, float green, float blue, float alpha)
void	SetBackgroundColor (float red, float green, float blue, float alpha)
void	SetParameters (float frequency, float persistence, int octaves, bool smooth)
void	SetInterpolationMode (int mode)
void	TransformXY (float(*processor)(float x, float y, float value))
void	TransformAB (float a, float b, float(*processor)(float a, float b, float value))
void	GenerateNoise ()
void	ExportRGB256 (GLubyte *data_pointer)
void	Export (float *data_pointer)
void	GetDimensions (int &dim_x, int &dim_y)
void	Push ()
void	FlushToColorBuffer ()
void	CombineColorBuffers (float(*combiner)(float val_1, float val_2))
void	CombineColorBuffersAB (float a, float b, float(*combiner)(float a, float b, float val_1, float val_2))
void	SmoothBorder (int border)
Static Public Member Functions
static float	RangeTransform (float a, float b, float perlin_xy)
static float	LinearTransform (float a, float b, float perlin_xy)
static float	MarblyTransform (float x, float y, float perlin_xy)
static float	GrainyTransform (float x, float y, float perlin_xy)
static float	SumCombiner (float val_1, float val_2)
static float	ProductCombiner (float val_1, float val_2)
static float	MaximumCombiner (float val_1, float val_2)
static float	WeightedSumCombiner (float weight_1, float weight_2, float val_1, float val_2)
Private Types
typedef PerlinNoise::StackElem	PerlinStack
Private Member Functions
	PerlinNoise ()
void	Pop ()
bool	IsEmpty ()
float	GetSmoothedValue (float x, float y)
float	GetNoiseValue (float x, float y)
float	GetNoise2D (int x, int y)
float	Interpolate (float value_1, float value_2, float fraction)
float	LinearInterpolation (float value_1, float value_2, float fraction)
float	CosineInterpolation (float value_1, float value_2, float fraction)
float	CubicInterpolation (float value_1, float value_2, float fraction)
Private Attributes
int	size_x
int	size_y
PerlinStack *	top
Classes
struct	StackElem

---

Detailed Description

Class used to generate 2D perlin noise.

PerlinNoise is a class that can be used to generate 2 dimensional perlin noise. It contains some modified, rewritten code from the two internet sources listed below.

The whole class is stack based, which means that additional elements can be pushed onto the stack and any operation like generating the noise, transforming the noise or setting parameters is only applied to the top of the stack. Furthermore, there are some combination routines, which combine the top two elements of the stack into one single element and thereby reduce the stack size by one.

Every stack element contains two buffers, a buffer with the actual noise and a color buffer. The actual noise buffer can be filled by generating noise (with some parameters). This buffer can then be modified by applying certain transformation routines to each value of the buffer. This buffer can either be exported directly or be written into the color buffer of the stack element. This procedure will generate color values from the noise values by interpolating between the foreground and background color attributes. The color buffers of the stack elements can then either be exported to an array of color values or combined with each other by applying a certain combination routine to each of the color values.

For transforming the noise in the noise buffer or combining two color buffers of the stack, there are some basic routines implemented as static members of this class. But the transformations and combinations are in no way restriced to these routines, any routine can be used for transforming and combining the noise as long as it has the correct argument and return value types.

Sources: - http://www.animeimaging.com/asp/PerlinNoise.aspx

• http://freespace.virgin.net/hugo.elias/models/m_perlin.htm

Definition at line 46 of file PerlinNoise.h.

---

Member Typedef Documentation

typedef struct PerlinNoise::StackElem PerlinNoise::PerlinStack [private]

Structure holding an element of the stack.

---

Constructor & Destructor Documentation

PerlinNoise::PerlinNoise (  int  size_x, int  size_y )

Constructor: Brings the object into a consistent state. The dimensions of the perlin noise has to be specified when the object is created. Parameters: size_x  x dimension of the perlin noise. size_y  y dimension of the perlin noise. Definition at line 3 of file PerlinNoise.cpp.

PerlinNoise::~PerlinNoise (   )

Destructor. Definition at line 14 of file PerlinNoise.cpp.

PerlinNoise::PerlinNoise (   )  [private]

Hide the default constructor.

---

Member Function Documentation

void PerlinNoise::CombineColorBuffers (  float(*)(float val_1, float val_2)  combiner  )

Combines the values of the color buffers of the 2 topmost elements on the stack by applying a function to all the values in the color buffer. After the two color buffers have been combined, the top element of the stack is popped from the stack. The resulting color buffer will then be on top of the stack. Parameters: combiner  Pointer to a function that takes to values as input and returns an output value. This function will be applied to all the values of the two color buffers. Definition at line 195 of file PerlinNoise.cpp.

void PerlinNoise::CombineColorBuffersAB (  float  a, float  b, float(*)(float a, float b, float val_1, float val_2)  combiner )

Combines the values of the color buffers of the 2 topmost elements on the stack by applying a function that takes two additional parameters to all the values in the color buffer. After the two color buffers have been combined, the top element of the stack is popped from the stack. The resulting color buffer will then be on top of the stack Parameters: a  First parameter that shall be fed to the combiner function. b  Second parameter that shall be fed to the combiner function. combiner  Pointer to the combiner function. This function will be fed with both parameters as well as the two values of the color buffers and it must return the resulting color value. Definition at line 214 of file PerlinNoise.cpp.

float PerlinNoise::CosineInterpolation (  float  value_1, float  value_2, float  fraction )  [private]

Use cosine interpolation between two values. Parameters: value_1  First value for the interpolation. value_2  Second value for the interpolation. fraction  Fraction between 0 and 1 used for the interpolation. (0: first value, 1: second value). Definition at line 336 of file PerlinNoise.cpp.

float PerlinNoise::CubicInterpolation (  float  value_1, float  value_2, float  fraction )  [private]

Use cubic interpolation between two values. Parameters: value_1  First value for the interpolation. value_2  Second value for the interpolation. fraction  Fraction between 0 and 1 used for the interpolation. (0: first value, 1: second value). Definition at line 342 of file PerlinNoise.cpp.

void PerlinNoise::Export (  float *  data_pointer  )

Exports the values of the generated perlin noise on top of the stack to an array of single precision floating point values. Parameters: data_pointer  pointer to an array of single precision floating point values. This array will be filled with the perlin noise values and its size has to be of at least size_x * size_y elements. Definition at line 187 of file PerlinNoise.cpp.

void PerlinNoise::ExportRGB256 (  GLubyte *  data_pointer  )

Exports the values of the color buffer on top of the stack to an array of unsigned bytes in RGB format. This routine does not produce any meaningful result if 'FlushToColorBuffer ()' has not been called on the element currently on top of the stack. Parameters: data_pointer  pointer to an array of unsigned bytes. This array will be filled with color values from 0 to 255 in RGB format and its size has to be of at least size_x * size_y * 3 elements. Definition at line 172 of file PerlinNoise.cpp.

void PerlinNoise::FlushToColorBuffer (   )

Flushes the perlin noise to the color buffer by linearly interpolating the values of the perlin noise between the background and foreground color. Because the values of the perlin noise may be smaller than 0 or bigger than 1, there is no guarantee that the values written to the color buffer are between the background and foreground color. This can, however, be achieved by appying a range transform (0..1) on the perlin noise before flushing it to the color buffer. This routine does only affect the top of the stack. Definition at line 67 of file PerlinNoise.cpp.

void PerlinNoise::GenerateNoise (   )

Generates a two dimensional perlin noise using the actual parameters. Transformations, direct exports and flushed to the color buffer do not produce meaningful results unless this routine has been called first. This routine does only affect the top of the stack. Definition at line 111 of file PerlinNoise.cpp.

void PerlinNoise::GetDimensions (  int &  dim_x, int &  dim_y )

Returns the x and y dimension of the perlin noise. Parameters: dim_x  Variable passed by reference. After the routine, this value will contain the size of the perlin noise in the x dimension. dim_y  Variable passed by reference. After the routine, this value will contain the size of the perlin noise in the y dimension. Definition at line 105 of file PerlinNoise.cpp.

float PerlinNoise::GetNoise2D (  int  x, int  y )  [private]

Gets the noise at a discrete location. Parameters: x  Discrete x coordinate for which the noise value shall be retrieved. y  Discrete y coordinate for which the noise value shall be retrieved. Definition at line 303 of file PerlinNoise.cpp.

float PerlinNoise::GetNoiseValue (  float  x, float  y )  [private]

Retrives noise at one specific point. Parameters: x  X coordinate for which the noise value shall be retrieved. y  Y coordinate for which the noise value shall be retrieved. Definition at line 278 of file PerlinNoise.cpp.

float PerlinNoise::GetSmoothedValue (  float  x, float  y )  [private]

Retrieves a smoothed noise value at one specific point. This is done by evaluating the noise at (x-1,y-1), (x,y-1), (x+1,y-1), (x-1,y), (x,y), (x+1,y-1), (x-1,y+1), (x,y+1), (x+1,y+1) and weighting these values in a meaningful way. Parameters: x  X coordinate for which the smoothed noise value shall be retrieved. y  Y coordinate for which the smoothed noise value shall be retrieved. Definition at line 265 of file PerlinNoise.cpp.

float PerlinNoise::GrainyTransform (  float  x, float  y, float  perlin_xy )  [static]

Predefined transformation function that can be used with 'TransformXY (...)'. Makes the noise appear grainy. Parameters: x  X coordinate at which the input value shall be transformed. y  Y coordinate at which the input value shall be transformed. perlin_xy  Input value. Definition at line 367 of file PerlinNoise.cpp.

float PerlinNoise::Interpolate (  float  value_1, float  value_2, float  fraction )  [private]

Interpolates between two values using the interpolation specified by the interpolation mode. Parameters: value_1  First value for the interpolation. value_2  Second value for the interpolation. fraction  Fraction between 0 and 1 used for the interpolation. (0: first value, 1: second value). Definition at line 315 of file PerlinNoise.cpp.

bool PerlinNoise::IsEmpty (   )  [private]

Checks wheather there are elements on the stack or not. Definition at line 62 of file PerlinNoise.cpp.

float PerlinNoise::LinearInterpolation (  float  value_1, float  value_2, float  fraction )  [private]

Interpolate linearly between two values. Parameters: value_1  First value for the interpolation. value_2  Second value for the interpolation. fraction  Fraction between 0 and 1 used for the interpolation. Definition at line 331 of file PerlinNoise.cpp.

float PerlinNoise::LinearTransform (  float  a, float  b, float  perlin_xy )  [static]

Predefined transformation function that can be used with 'TransformAB (...)'. Transforms each point independently with new_value = old_value * a + b. Parameters: a  Input value will be multiplied by this value. b  Offset that will be added to the input value. perlin_xy  Input value. Definition at line 357 of file PerlinNoise.cpp.

float PerlinNoise::MarblyTransform (  float  x, float  y, float  perlin_xy )  [static]

Predefined transformation function that can be used with 'TransformXY (...)'. Makes the noise appear marbly. Parameters: x  X coordinate at which the input value shall be transformed. y  Y coordinate at which the input value shall be transformed. perlin_xy  Input value. Definition at line 362 of file PerlinNoise.cpp.

float PerlinNoise::MaximumCombiner (  float  val_1, float  val_2 )  [static]

Predefined combiner function that can be used with 'CombineColorBuffers'. Returns the maximum of the two input values. Parameters: val_1  Value from the first buffer. val_2  Value from the second buffer. Returns: The maximum of both values. Definition at line 385 of file PerlinNoise.cpp.

void PerlinNoise::Pop (   )  [private]

Pops the top element from the stack. Internally used after combining color buffers. Definition at line 46 of file PerlinNoise.cpp.

float PerlinNoise::ProductCombiner (  float  val_1, float  val_2 )  [static]

Predefined combiner function that can be used with 'CombineColorBuffers'. Returns the product of the two input values. Parameters: val_1  Value from the first buffer. val_2  Value from the second buffer. Returns: The product of both values. Definition at line 380 of file PerlinNoise.cpp.

void PerlinNoise::Push (   )

Pushes a new element onto the stack. Definition at line 20 of file PerlinNoise.cpp.

float PerlinNoise::RangeTransform (  float  a, float  b, float  perlin_xy )  [static]

Predefined transformation function that can be used with 'TransformAB (...)'. Makes sure that the resulting value lies within a certain interval. Parameters: a  Lower bound of the interval, all returned values will be at least as big as this parameter. b  Upper bound of the interval, all returned values will be smaller or equal to this parameter. perlin_xy  Input value. Definition at line 349 of file PerlinNoise.cpp.

void PerlinNoise::SetBackgroundColor (  float  red, float  green, float  blue, float  alpha )

Sets the background color that will be used when the noise is transformed into color space. This routine does only affect the top of the stack. Parameters: red  Red part of the background color. green  Green part of the background color. blue  Blue part of the background color. alpha  Alpha value of the background color. Definition at line 76 of file PerlinNoise.cpp.

void PerlinNoise::SetForegroundColor (  float  red, float  green, float  blue, float  alpha )

Sets the foreground color that will be used when the noise is transformed into color space. This routine does only affect the top of the stack. Parameters: red  Red part of the foreground color. green  Green part of the foreground color. blue  Blue part of the foreground color. alpha  Alpha value of the foreground color. Definition at line 84 of file PerlinNoise.cpp.

void PerlinNoise::SetInterpolationMode (  int  mode  )

Sets the mode of interpolation used between the bare random noise values. This routine does only affect the top of the stack. Parameters: mode  Interpolation mode (LINEAR_INTERPOLATION, COSINE_INTERPOLATION or CUBIC_INTERPOLATION). Definition at line 100 of file PerlinNoise.cpp.

void PerlinNoise::SetParameters (  float  frequency, float  persistence, int  octaves, bool  smooth )

Sets the parameters used for the noise generation. This routine does only affect the top of the stack. Parameters: frequency  Lowest frequency of the noise. This frequency is independent of the size of the noise. persistence  Persistence defines how much higher frequencies are weighted with respect to the lower frequencies. octaves  Number of noise patterns that are added to the final noise. smooth  Flag indicating whether the noise shall be smoothed or not. Definition at line 92 of file PerlinNoise.cpp.

void PerlinNoise::SmoothBorder (  int  border  )

Smoothes the noise at the border such by linearly interpolating the values that are not farther away from the border that a certain distance. All values on the border are will be zero afterwards. Parameters: border  Width of border in which the values are interpolated linearly. Definition at line 233 of file PerlinNoise.cpp.

float PerlinNoise::SumCombiner (  float  val_1, float  val_2 )  [static]

Predefined combiner function that can be used with 'CombineColorBuffers'. Returns the sum of the two input values. Parameters: val_1  Value from the first buffer. val_2  Value from the second buffer. Returns: The sum of both values. Definition at line 375 of file PerlinNoise.cpp.

void PerlinNoise::TransformAB (  float  a, float  b, float(*)(float a, float b, float value)  processor )

Transforms the generated noise with a function that takes two parameters a and b. Each value of the noise is transformed by calling the function with both parameters on each value of the perlin noise. This routine does only affect the top of the stack. Parameters: a  First parameter that will be fed to the function specified as the third argument of this routine. b  Second parameter that will be fed to the function specified as the third argument of this routine. processor  Pointer to a function which takes two parameter as well as an input value and returns a transformed value. Definition at line 164 of file PerlinNoise.cpp.

void PerlinNoise::TransformXY (  float(*)(float x, float y, float value)  processor  )

Transforms the generated noise with a function that depends on the x and y coordinated of the noise. x and y, which are fed to this function, are not the dimensions of the noise. They merely depend on the frequency times coordinate values between 0 and 1 (0 at top and left/ 1 at bottom and right). This routine does only affect the top of the stack. Parameters: processor  Pointer to a function which takes an x and y coordinate as well as an input value and returns a transformed value. Definition at line 149 of file PerlinNoise.cpp.

float PerlinNoise::WeightedSumCombiner (  float  weight_1, float  weight_2, float  val_1, float  val_2 )  [static]

Predefined combiner function that can be used with 'CombineColorBuffersAB'. Returns the weighted sum of the two input values. Parameters: weight_1  The weight for the value from the first buffer. weight_2  The weight for the value from the second buffer. val_1  Value from the first buffer. val_2  Value from the second buffer. Returns: The weighted sum of both values (weight_1 * val_1 + weight_2 * val_2). Definition at line 390 of file PerlinNoise.cpp.

---

Member Data Documentation

int PerlinNoise::size_x [private]

Size of the noise along the x dimension. Definition at line 300 of file PerlinNoise.h.

int PerlinNoise::size_y [private]

Size of the noise along the y dimention. Definition at line 303 of file PerlinNoise.h.

PerlinStack* PerlinNoise::top [private]

Top element of the stack. Definition at line 322 of file PerlinNoise.h.

---