ihsanenginbal/earthquake_wavelets

We produced RGB pictures where the X axis is the time in seconds, the Y axis is the frequency of the waves and the color intensity represents the energy content of the wave at that moment, within that frequency.

The wavelets represent 120sec long records.

This dataset has wavelet pictures from earthquakes, from stormy days, from rush hours as well as from sleepy hours during a day.

In order to better understand how the wavelets are produced, you can check the below code with which we produced the wavelets.

import pywt import numpy as np import matplotlib.pyplot as plt from matplotlib import cm from pywt import scale2frequency import math from scipy.io import loadmat from joblib import delayed, Parallel from tqdm import tqdm from matplotlib.pyplot import figure

def plot(data, fig_index): # parmaters for cwt # deltat: float = 0.005 wavelet = 'morl'

per1 = 1 / 20  # minimum Frequency
per2 = 1 / 0.1  # minimum Frequency

cfreq = 0.8125  # default value for morl cwt #

scale1 = per1 * cfreq / deltat
scale2 = per2 * cfreq / deltat

scales = np.arange(scale1, scale2 + (scale2 - scale1) / 15, (scale2 - scale1) / 15)
coefs, freqs = pywt.cwt(data, scales, wavelet)  # python cwt function

coeflist = np.array(coefs)  # to convert from tuple to ndarray

S = np.sqrt(np.absolute(coeflist))
SC = (100 * S) / np.sum(np.array(S).flatten())
F = pywt.scale2frequency(wavelet, scales) / deltat

time = np.arange(deltat, (len(data) * deltat) + deltat, deltat)

plt.contourf(time, 1 / F, SC, cmap=cm.jet)

#     scrsz = [100,100,1000,700];
#     plt.figure(figsize=(100, 200, scrsz(2)*0.80, scrsz(3)*0.6))
plt.axis('off')
plt.savefig('Storm_Plots/contourf_' + str(fig_index) + '.png', dpi=500)
# plt.show()

#### plot screen size should be fixed ###

def process(record, index): plot(record.flatten(), fig_index=index + 1)

EQ = loadmat('Storm_Data_Katerina_All.mat') records=EQ['Storm_Data_Katerina_All']['recData'] print(len(records[0]))

Parallel(n_jobs=10)(delayed(process)(record, index) for index, record in enumerate(tqdm(records[0])))