Figure 4 script
Figure 4 scriptΒΆ
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd
import seaborn as sns
import matplotlib as mpl
# font and linewidth settings
font = {'size' : 20}
mpl.rc('font', **font)
mpl.rc('legend', fontsize=16)
axes_settings = {'linewidth' : 2}
mpl.rc('axes', **axes_settings)
Now read the data:
df = pd.read_csv('../results_2022.csv', dtype=str)
df = df.replace('varying: multi-pass', 0) # replace some values in 'Template size (m)' with another flag
for field in ['Pixel spacing (m)',
'Template size (m)',
'SAV-uncertainty-x',
'SAV-uncertainty-y',
'SAV-peak-x',
'SAV-peak-y',
'LSR-uncertainty-nm',
'LSR-uncertainty-sh',
'pt0_vxdiff',
'pt0_vydiff',
'pt1_vxdiff',
'pt1_vydiff',
'pt2_vxdiff',
'pt2_vydiff',
'pt0_vxavgdiff',
'pt0_vyavgdiff',
'pt1_vxavgdiff',
'pt1_vyavgdiff',
'pt2_vxavgdiff',
'pt2_vyavgdiff',
'SAV-outlier-percent',
'Invalid-pixel-percent']:
df[field] = df[field].astype(float)
datestrs = ['LS8-20180304-20180405',
'LS8-20180802-20180818',
'Sen2-20180304-20180314',
'Sen2-20180508-20180627']
# df
Now plot the figure:
demo1 = df[df['Prefilter'] != 'LoG']
demo2 = df[df['Template size (px)'] != "48"]
demo3 = df[df['Pixel spacing (px)'] != "12"]
for idx, row in demo2.iterrows():
if row['Template size (px)'] in ["64", "65"]:
demo2.loc[idx, 'Template size (pixels)'] = "64-65"
elif row['Template size (px)'] in ["31", "32"]:
demo2.loc[idx, 'Template size (pixels)'] = "31-32"
elif row['Template size (px)'] == 0:
demo2.loc[idx, 'Template size (pixels)'] = "multi"
else:
demo2.loc[idx, 'Template size (pixels)'] = "else"
for idx, row in demo3.iterrows():
if row['Pixel spacing (px)'] == "1":
demo3.loc[idx, 'Pixel spacing (pixels)'] = "1"
elif row['Pixel spacing (px)'] in ["4", "4.009", "4.003"]:
demo3.loc[idx, 'Pixel spacing (pixels)'] = "~4"
elif row['Pixel spacing (px)'] == "8":
demo3.loc[idx, 'Pixel spacing (pixels)'] = "8"
elif row['Pixel spacing (px)'] in ["15.13", "16.04"]:
demo3.loc[idx, 'Pixel spacing (pixels)'] = "~16"
else:
demo3.loc[idx, 'Pixel spacing (pixels)'] = "else"
fig, axs = plt.subplots(3, 1, figsize=(13, 11), constrained_layout=True)
kawgs = {'jitter': 0.15, 'marker': 'D', 's': 10, 'alpha': 0.3, 'linewidth': 2, }
sns.stripplot(data=demo1, x="SAV-uncertainty-x", y="Prefilter", ax=axs[0],
order=['None', 'Gau', 'NAOF'], **kawgs)
sns.stripplot(data=demo2, x="SAV-uncertainty-x", y="Template size (pixels)", ax=axs[1],
order=['multi', '31-32', '64-65'], **kawgs)
sns.stripplot(data=demo3, x="LSR-uncertainty-sh", y="Pixel spacing (pixels)", ax=axs[2], **kawgs)
kawgs2 = {'linewidth': 3, }
tmp = demo1[demo1['Prefilter'] == 'None']
tmp_mean = tmp['SAV-uncertainty-x'].median()
axs[0].plot([tmp_mean, tmp_mean], [-0.3, 0.3], **kawgs2)
tmp = demo1[demo1['Prefilter'] == 'Gau']
tmp_mean = tmp['SAV-uncertainty-x'].median()
axs[0].plot([tmp_mean, tmp_mean], [0.7, 1.3], **kawgs2)
tmp = demo1[demo1['Prefilter'] == 'NAOF']
tmp_mean = tmp['SAV-uncertainty-x'].median()
axs[0].plot([tmp_mean, tmp_mean], [1.7, 2.3], **kawgs2)
axs[0].set_xlabel('$\delta_u$ (m day$^{-1}$)')
tmp = demo2[demo2['Template size (pixels)'] == 'multi']
tmp_mean = tmp['SAV-uncertainty-x'].median()
axs[1].plot([tmp_mean, tmp_mean], [-0.3, 0.3], **kawgs2)
tmp = demo2[demo2['Template size (pixels)'] == '31-32']
tmp_mean = tmp['SAV-uncertainty-x'].median()
axs[1].plot([tmp_mean, tmp_mean], [0.7, 1.3], **kawgs2)
tmp = demo2[demo2['Template size (pixels)'] == '64-65']
tmp_mean = tmp['SAV-uncertainty-x'].median()
axs[1].plot([tmp_mean, tmp_mean], [1.7, 2.3], **kawgs2)
axs[1].set_xlabel('$\delta_u$ (m day$^{-1}$)')
tmp = demo3[demo3['Pixel spacing (pixels)'] == '1']
tmp_mean = tmp['LSR-uncertainty-sh'].median()
axs[2].plot([tmp_mean, tmp_mean], [-0.3, 0.3], **kawgs2)
tmp = demo3[demo3['Pixel spacing (pixels)'] == '~4']
tmp_mean = tmp['LSR-uncertainty-sh'].median()
axs[2].plot([tmp_mean, tmp_mean], [0.7, 1.3], **kawgs2)
tmp = demo3[demo3['Pixel spacing (pixels)'] == '8']
tmp_mean = tmp['LSR-uncertainty-sh'].median()
axs[2].plot([tmp_mean, tmp_mean], [1.7, 2.3], **kawgs2)
tmp = demo3[demo3['Pixel spacing (pixels)'] == '~16']
tmp_mean = tmp['LSR-uncertainty-sh'].median()
axs[2].plot([tmp_mean, tmp_mean], [2.7, 3.3], **kawgs2)
axs[2].set_xlabel("$\delta_{x'y'}$ (day$^{-1}$)")
axs[1].set_yticklabels(['4-24\n[GIV]', '31-32', '64-65'])
axs[2].set_yticklabels(['1', '4', '8', '15-16'])
axs[0].set_xlim(0, 1)
axs[1].set_xlim(0, 1)
axs[2].set_xlim(0, 0.06)
axs[0].text(-0.11, 1, "$\mathbf{a}$", transform=axs[0].transAxes)
axs[1].text(-0.11, 1, "$\mathbf{b}$", transform=axs[1].transAxes)
axs[2].text(-0.11, 1, "$\mathbf{c}$", transform=axs[2].transAxes)
axs[1].set_ylabel("Template size \n (pixels)")
axs[2].set_ylabel("Output resolution \n (pixels)")
# save figure
fig.patch.set_facecolor('xkcd:white')
fig.savefig('Fig4.png', dpi=200)
