NOISE_PERIODIC Function
Generates an array of periodic (coherent) noise.
Usage
result = NOISE_PERIODIC(d1[, d2, ... , d8])
Input Parameters
d1, ..., d8—The dimensions of the noise array.
Returned Value
result—An array containing periodic noise. The array data type is specified using the Type keyword.
Keywords
Amp—A scalar float that is the maximum amplitude of the periodic noise. (Default: 1.0)
DC_Offset—A scalar float containing the DC offset (also known as the zero frequency) of the periodic noise. (Default: 0.0)
F1, ..., F8—The spatial frequency of the noise in dimensions d1, ..., d8, respectively. (Default: 0.0)
Type—A scalar string specifying the data type of the returned noise array. Valid strings are: 'byte', 'fix', 'long', 'float', and 'double'. (Default: 'float')
Discussion
Periodic, or coherent, noise is composed of 2D sinusoidal functions. This is often seen in the form of electrical noise at 60 (or 50) Hz. In the spatial frequency domain, periodic noise corruption is easily seen as bright spot in the spectrum. A notch filter is generally useful in eliminated periodic noise from an image.
The formula for periodic noise is as follows:
n(i, j) = A sin(f0i + f1j)
Example
; First, get an array of periodic noise.
noise = NOISE_PERIODIC(128, 128, F1 = 30.0, $
F2 = 100.0, Amp = 15.0, DC_Offset = 128.0)
; Now, corrupt an image with the generated periodic noise.
corrupt_image = IPMATH(image, '+', noise, /No_Clip)
See Also
