PeakUtils tutorial

This tutorial shows the basic usage of PeakUtils to detect the peaks of 1D data.

Importing the libraries



In [7]:

    
import numpy
import peakutils
from peakutils.plot import plot as pplot
from matplotlib import pyplot
%matplotlib inline

Preparing the data

Now, lets generate some noisy data with two gaussians:



In [2]:

    
centers = (30.5, 72.3)
x = numpy.linspace(0, 120, 121)
y = peakutils.gaussian(x, 5, centers[0], 3) + peakutils.gaussian(x, 7, centers[1], 10) + numpy.random.rand(x.size)
pyplot.figure(figsize=(10,6))
pyplot.plot(x, y)
pyplot.title("Data with noise")









    Out[2]:





<matplotlib.text.Text at 0x7f5eaec6d278>

Getting a first estimate of the peaks

By using peakutils.indexes, we can get the indexes of the peaks from the data. Due to noise, it will be just a rough approximation.



In [8]:

    
indexes = peakutils.indexes(y, thres=0.5, min_dist=30)
print(indexes)
print(x[indexes], y[indexes])
pyplot.figure(figsize=(10,6))
pplot(x, y, indexes)
pyplot.title('First estimate')









    



[31 74]
[ 31.  74.] [ 5.67608909  7.79403394]






    Out[8]:





<matplotlib.text.Text at 0x7f5eaebf6198>

Enhancing the resolution by interpolation

We can enhance the resolution by interpolation. Here, we will try to fit a Gaussian near each peak that we have just detected.



In [9]:

    
peaks_x = peakutils.interpolate(x, y, ind=indexes)
print(peaks_x)









    



[ 30.58270223  72.34348214]

Estimating and removing the baseline

It is common for data to have a baseline that may not be desired. PeakUtils implements one function for estimating the baseline by using an iterative polynomial regression algorithm:



In [10]:

    
y2 = y + numpy.polyval([0.002,-0.08,5], x)
pyplot.figure(figsize=(10,6))
pyplot.plot(x, y2)
pyplot.title("Data with baseline")









    Out[10]:





<matplotlib.text.Text at 0x7f5eaeb73400>



In [11]:

    
base = peakutils.baseline(y2, 2)
pyplot.figure(figsize=(10,6))
pyplot.plot(x, y2-base)
pyplot.title("Data with baseline removed")









    Out[11]:





<matplotlib.text.Text at 0x7f5eae14e940>

Scipy also implements functions that can be used for peak detection. Some examples:



In [ ]: