burst_imaging/find_maxburst.py at master · murphp30/burst_imaging · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
#!/usr/bin/env python
import matplotlib
matplotlib.use('TkAgg')

import argparse

import astropy.units as u
import matplotlib.gridspec as gridspec
import matplotlib.pyplot as plt
plt.rcParams.update({'font.size':14})
import numpy as np
import pandas as pd

from matplotlib import dates
from scipy.ndimage import gaussian_filter1d, uniform_filter1d
from scipy.signal import savgol_filter, find_peaks
from sunpy.time import TimeRange

from LOFAR_bf import LOFAR_BF


def rolling_window(a, window):
    shape = a.shape[:-1] + (a.shape[-1] - window + 1, window)
    strides = a.strides + (a.strides[-1],)
    return np.lib.stride_tricks.as_strided(a, shape=shape, strides=strides)


def rebin_array(arr, new_index):
    return np.array(
        [np.sum(arr[i * new_index:(i + 1) * new_index]) for i in range(int(np.floor(len(arr) / new_index)))])


parser = argparse.ArgumentParser(description='Find a number of peaks in a LOFAR LTA dynamic spectrum')
parser.add_argument('f', help='filename of h5 LTA file.', metavar='FILENAME')
parser.add_argument('-t', '--trange', nargs=2,
                    help='time range for observation.2 arguments START and END in format YYYY-MM-DDTHH:MM:SS',
                    metavar=('START', 'END'))
parser.add_argument('-p', '--plot',
                    help='plot dynamic spectrum. Default is True',
                    action='store_true')
args = parser.parse_args()
f = args.f
trange = args.trange
plot = args.plot

if trange is not None:
    trange = TimeRange(trange[0], trange[1])  # ("2019-04-04T14:08:00", "2019-04-04T14:17:00")
clip = u.Quantity([5, 95] * u.percent)

bf = LOFAR_BF(f, trange)
t0 = int(np.floor((bf.trange.start - bf.obs_start).sec / bf.dt.sec))
t1 = int(np.floor((bf.trange.end - bf.obs_start).sec / bf.dt.sec))
# tarr = np.arange(bf.data.shape[0]) * bf.dt
# tarr = tarr + bf.trange.start
tarr = bf.obs_start+np.arange(t0,t1+1)*bf.dt
new_dt = 0.167772  # 0.167772 seconds is interferometric temporal resolution
new_dt_index = int(np.round(new_dt / bf.dt.sec))
sig = new_dt_index / (2 * np.sqrt(2 * np.log(2)))  # FWHM is interferometric time res.
freqs = [30.4687500]#[51.3671875]  # np.arange(20,80,10)
if plot:
    fig, ax = plt.subplots(figsize=(10, 8))
    bf.plot(ax=ax, bg_subtract=True, clip_interval=clip)
peaks_df = []#pd.DataFrame()
max_peaks_df = []
for freq in freqs:
    loc = np.where(abs(bf.freqs.to(u.MHz) - freq * u.MHz) == np.min(abs(bf.freqs.to(u.MHz) - freq * u.MHz)))[0][0]
    dslice = np.mean(bf.data[:, loc - 8:loc + 8], axis=1)  # bf.data[:, loc] average over 16 channels
    dslice = rebin_array(dslice, new_dt_index)
    tarr = tarr[new_dt_index//2:-(len(tarr)%new_dt_index):new_dt_index]#tarr[new_dt_index//2::new_dt_index]
    smooth = dslice#gaussian_filter1d(dslice, sig)
    # define a background using rolling window of standard deviations
    if trange.dt.sec > 60:
        win_len = int(np.round(10 / bf.dt.sec))  # 10 second window
    else:
        win_len = int(np.round(1 / bf.dt.sec))  # 1 second window
    stds = np.std(rolling_window(smooth, win_len), 1)
    bg_std = stds[np.where(stds == np.min(stds))[0][0]]
    means = np.mean(rolling_window(smooth, win_len), 1)
    bg_mean = means[np.where(stds == np.min(stds))[0][0]]

    if trange.dt.sec > 60:
        peaks, _ = find_peaks(smooth,
                          height= bg_mean + 5 * bg_std,
                          distance= 10 * new_dt_index,
                          prominence= 10*bg_std)
    else:
        peaks, _ = find_peaks(smooth,
                              height=bg_mean + bg_std)
    # width = new_dt_index)
    # prominence=(np.max(smooth)-np.mean(smooth))/10)
    # peaks, _ = find_peaks(dslice,
    #        height=np.mean(dslice)+3*np.std(dslice))
    # prominence=(np.max(smooth)-np.mean(smooth))/10)

    # ax.scatter(tarr.plot_date[peaks],
    #         np.ones(len(peaks))*bf.freqs[loc].to(u.MHz).value,
    #         color='r', marker='+')
    # max_peak = peaks[np.where(dslice[peaks] == np.max(dslice[peaks]))[0][0]]
    max_peak = peaks[np.where(smooth[peaks] == np.max(smooth[peaks]))[0][0]]
    max_peak_time = tarr[max_peak]
    peak_df = pd.DataFrame(tarr.isot[peaks], columns=[bf.freqs[loc].to(u.MHz).value])
    peaks_df.append(peak_df) #= peaks_df.append(peak_df)
    print('Maximum peak at {}'.format(max_peak_time))
    print('Bursts found: {}'.format(len(peaks)))
    max_peak_df = pd.DataFrame([max_peak_time.isot], columns=[bf.freqs[loc].to(u.MHz).value])
    max_peaks_df.append(max_peak_df)  # = peaks_df.append(peak_df)
    if plot:
        ax.scatter(tarr.plot_date[peaks],
                   np.ones(len(peaks)) * bf.freqs[loc].to(u.MHz).value,
                   color='w', marker='o')
        ax.scatter(tarr.plot_date[max_peak],
                   bf.freqs[loc].to(u.MHz).value,
                   color='r', marker='+')
        # ax.hlines([bf.freqs[loc-8].to(u.MHz).value, bf.freqs[loc+8].to(u.MHz).value], tarr.plot_date[0], tarr.plot_date[-1], color='r')

fig, ax = plt.subplots(figsize=(13, 10))
ax.plot(tarr.plot_date, dslice)
ax.plot(tarr.plot_date[peaks], dslice[peaks], 'o', color='r')
ax.set_yscale('log')
ax.xaxis_date()
date_format = dates.DateFormatter("%H:%M:%S")
ax.xaxis.set_major_formatter(date_format)
ax.set_ylabel('Intensity (arbitrary)')
ax.set_xlabel('Time (UTC)')
plt.cloes()
# # plt.plot(tarr.plot_date, smooth)
# ax.vlines(tarr.plot_date[max_peak], np.min(dslice), dslice[max_peak], color='r', zorder=1000)
# ax.xaxis_date()
# date_format = dates.DateFormatter("%H:%M:%S")
# ax.xaxis.set_major_formatter(date_format)
# plt.hlines(np.mean(dslice) + 5 * bg_std, 0, len(smooth), color='k', zorder=1000)
save_path = "./"  # "/mnt/murphp30_data/paper2"
save_png = save_path + "/peak_times_30MHz_{}_{}.png".format(trange.start.isot[:-4].replace(':', ''),
                                                      trange.end.isot[11:-4].replace(':', ''))
# if plot:
#     plt.savefig(save_png)
peaks_df = pd.concat(peaks_df, axis=1)
max_peaks_df = pd.concat(max_peaks_df, axis=1)
save_pickle = save_path + "/peak_times_30MHz_{}_{}.pkl".format(trange.start.isot[:-4].replace(':', ''),
                                                            trange.end.isot[11:-4].replace(':', ''))
# with open("peak_times.txt", 'a') as peak_file:
#     peak_file.write(max_peak_time.isot)
#     peak_file.write('\n')
# peaks_df.to_pickle(save_pickle)
# max_peaks_df.to_pickle(save_pickle)
# plt.close()
fig = plt.figure(figsize=(10, 8))
gs = gridspec.GridSpec(ncols=1, nrows=2, height_ratios=[1,2])
ax0 = fig.add_subplot(gs[1])
ax1 = fig.add_subplot(gs[0], sharex=ax0)
bf.plot(ax=ax0, bg_subtract=True, clip_interval=clip)
# ax0.hlines(bf.freqs[loc].to(u.MHz).value, tarr.plot_date[0], tarr.plot_date[-1], color='r')
ax0.scatter(tarr.plot_date[peaks],
           np.ones(len(peaks)) * bf.freqs[loc].to(u.MHz).value,
           color='w', marker='o')
ax0.scatter(tarr.plot_date[max_peak],
           bf.freqs[loc].to(u.MHz).value,
           color='r', marker='+')
ax1.plot(tarr.plot_date, dslice)
ax1.plot(tarr.plot_date[peaks], dslice[peaks], 'o', color='r')
ax1.plot(tarr.plot_date[max_peak], dslice[max_peak], '+', color='w')
ax1.set_yscale('log')
ax1.set_ylabel('Intensity (arbitrary)')
ax1.tick_params('x', labelbottom=False)
plt.tight_layout()
plt.savefig(save_path + "/peak_times_30MHz_wlightcurve_{}_{}.png".format(trange.start.isot[:-4].replace(':', ''),
                                                      trange.end.isot[11:-4].replace(':', '')))
plt.show()