Finding peaks in a 1-D signal in Julia. The implementation is based on find_peaks in SciPy.
julia>]
pkg> add FindPeaks1Dusing FindPeaks1D, ImageFiltering, Plots
n = 48001
s1 = ImageFiltering.Kernel.gaussian((1000,), (n,))
s2 = ImageFiltering.Kernel.gaussian((500,), (n,))
s = s1.parent/maximum(s1.parent) + 0.5 * circshift(
s2.parent/maximum(s2.parent), (10000,))pkindices, properties = findpeaks1d(s;
height=0.1,
prominence=0.2,
width=1000.0,
relheight=0.9)
plot(s; color="black", label=false)
scatter!(pkindices, s[pkindices]; color="red", markersize=5, label="peaks")
vline!(properties["leftips"]; color="blue", width=2, label="peak edges")
vline!(properties["rightips"]; color="blue", width=2, label=false)
xlabel!("Sample")
pkindices, properties = findpeaks1d(s;
height=0.1,
distance=12000,
prominence=0.2,
width=1000.0,
relheight=0.9)
plot(s; color="black", label=false)
scatter!(pkindices, s[pkindices]; color="red", markersize=5, label="peaks")
vline!(properties["leftips"]; color="blue", width=2, label="peak edges")
vline!(properties["rightips"]; color="blue", width=2, label=false)
xlabel!("Sample")