def add_to_array(arr, area, time, bufr, savedir='/tmp', name=None, mode='L',
resize=None, ptsize=None, save=False):
"""Add synoptical reports from stations to provided geolocated image array
"""
# Create array image
arrshp = arr.shape[:2]
print(np.nanmin(arr), np.nanmax(arr))
arr_img = Image(arr, mode=mode)
# arr_img = Image(channels=[arr[:, :, 0], arr[:, :, 1], arr[:, :, 2]],
# mode='RGB')
arr_img.stretch('crude')
arr_img.invert()
arr_img.colorize(maskcol)
arr_img.invert()
# Import bufr
stations = read_synop(bufr, 'visibility')
currentstations = stations[time.strftime("%Y%m%d%H0000")]
lats = [i[2] for i in currentstations]
lons = [i[3] for i in currentstations]
vis = [i[4] for i in currentstations]
# Create array for synop parameter
visarr = np.empty(arrshp)
visarr.fill(np.nan)
x, y = (area.get_xy_from_lonlat(lons, lats))
vis_ma = np.ma.array(vis, mask=x.mask)
if ptsize:
xpt = np.array([])
ypt = np.array([])
for i, j in zip(x, y):
xmesh, ymesh = np.meshgrid(np.linspace(i - ptsize, i + ptsize,
ptsize * 2 + 1),
np.linspace(j - ptsize, j + ptsize,
ptsize * 2 + 1))
xpt = np.append(xpt, xmesh.ravel())
ypt = np.append(ypt, ymesh.ravel())
vispt = np.ma.array([np.full(((ptsize * 2 + 1,
ptsize * 2 + 1)), p) for p in vis_ma])
visarr[ypt.astype(int), xpt.astype(int)] = vispt.ravel()
else:
visarr[y.compressed(), x.compressed()] = vis_ma.compressed()
visarr_ma = np.ma.masked_invalid(visarr)
station_img = Image(visarr_ma, mode='L')
station_img.colorize(vis_colset)
station_img.merge(arr_img)
if resize is not None:
station_img.resize((arrshp[0] * resize, arrshp[1] * resize))
if name is None:
timestr = time.strftime("%Y%m%d%H%M")
name = "fog_filter_example_stations_{}.png".format(timestr)
if save:
savepath = os.path.join(savedir, name)
station_img.save(savepath)
return(station_img)
def add_to_image(image,
area,
time,
bufr,
savedir='/tmp',
name=None,
bgimg=None,
resize=None,
ptsize=None,
save=False):
"""Add synoptical visibility reports from station data to provided
geolocated image array
"""
arrshp = image.shape[:2]
# Add optional background image
if bgimg is not None:
# Get background image
bg_img = Image(bgimg.squeeze(), mode='L', fill_value=None)
bg_img.stretch("crude")
bg_img.convert("RGB")
# bg_img.invert()
image.merge(bg_img)
# Import bufr
stations = read_synop(bufr, 'visibility')
currentstations = stations[time.strftime("%Y%m%d%H0000")]
lats = [i[2] for i in currentstations]
lons = [i[3] for i in currentstations]
vis = [i[4] for i in currentstations]
# Create array for synop parameter
visarr = np.empty(arrshp)
visarr.fill(np.nan)
# Red - Violet - Blue - Green
vis_colset = Colormap((0, (228 / 255.0, 26 / 255.0, 28 / 255.0)),
(1000, (152 / 255.0, 78 / 255.0, 163 / 255.0)),
(5000, (55 / 255.0, 126 / 255.0, 184 / 255.0)),
(10000, (77 / 255.0, 175 / 255.0, 74 / 255.0)))
x, y = (area.get_xy_from_lonlat(lons, lats))
vis_ma = np.ma.array(vis, mask=x.mask)
if ptsize:
xpt = np.array([])
ypt = np.array([])
for i, j in zip(x, y):
xmesh, ymesh = np.meshgrid(
np.linspace(i - ptsize, i + ptsize, ptsize * 2 + 1),
np.linspace(j - ptsize, j + ptsize, ptsize * 2 + 1))
xpt = np.append(xpt, xmesh.ravel())
ypt = np.append(ypt, ymesh.ravel())
vispt = np.ma.array(
[np.full(((ptsize * 2 + 1, ptsize * 2 + 1)), p) for p in vis_ma])
visarr[ypt.astype(int), xpt.astype(int)] = vispt.ravel()
else:
visarr[y.compressed(), x.compressed()] = vis_ma.compressed()
visarr_ma = np.ma.masked_invalid(visarr)
station_img = Image(visarr_ma, mode='L')
station_img.colorize(vis_colset)
image.convert("RGB")
station_img.merge(image)
if resize is not None:
station_img.resize((arrshp[0] * resize, arrshp[1] * resize))
if name is None:
timestr = time.strftime("%Y%m%d%H%M")
name = "fog_filter_example_stations_{}.png".format(timestr)
if save:
savepath = os.path.join(savedir, name)
station_img.save(savepath)
return (station_img)
