powered by
Takes a matrix of numeric values and smoothes it by convolution with a symmetric Gaussian window function.
gaussianSmooth2D(m, kernelSize = 5, kernelSD = 0.5, plotKernel = FALSE)
input matrix (numeric, on any scale, doesn't have to be square)
the size of the Gaussian kernel, in points
the SD of the Gaussian kernel relative to its size (.5 = the edge is two SD's away)
if TRUE, plots the kernel
Returns a numeric matrix of the same dimensions as input.
modulationSpectrum
# NOT RUN { s = spectrogram(soundgen(), samplingRate = 16000, output = 'original', plot = FALSE) # image(log(s)) s1 = gaussianSmooth2D(s, kernelSize = 11, plotKernel = TRUE) # image(log(s1)) # }
Run the code above in your browser using DataLab