MASCDB camera-wise data#
The files MASCdb_CAM0.parquet,*MASCdb_CAM1.parquet*,*MASCdb_CAM2.parquet* contain the attributes listed in the table below.
Parameter |
Units |
Type |
Long name |
Reference / Format / Algorithm |
|---|---|---|---|---|
Global information |
||||
datetime |
datetime |
|||
campaign |
string |
Field campaign string |
||
pix_size |
m |
float |
Pixel size |
|
n_roi |
int |
# of ROIs
in raw image
|
Praz et al, 2017
Note: only one main ROI
is considered in the descriptors
|
|
cam_id |
int |
Camera ID: 0, 1, 2 |
||
roi_width |
pix |
int |
x-size of crop of
the selected ROI
|
|
roi_height |
pix |
int |
y-size of crop of
the selected ROI
|
|
roi_centroid_X |
pix |
int |
Centroid X coord
(original image)
|
|
roi_centroid_X |
pix |
int |
Centroid Y coord
( original image)
|
|
Particle size and area |
||||
area |
m^2 |
float |
Particle area |
Table A1:1 in Praz et al, 2017 |
perim |
m |
float |
Particle perimeter |
Table A1:2 in Praz et al, 2017 |
Dmean |
m |
float |
Mean (x/y) dimension |
Table A1:3 in Praz et al, 2017 |
Dmax |
m |
float |
Max dimension |
Table A1:4 in Praz et al, 2017 |
Dmax_ori |
deg |
float |
Orientation of Dmax |
|
Dmax_90 |
m |
float |
Max. dimension in
orthogonal dir.
of Dmax.
|
Baker and Lawson 2006 |
D90_r |
float |
|||
eq_radius |
m |
float |
Equiv. area radius |
Table A1:5 in Praz et al, 2017 |
area_porous |
m^2 |
float |
Porous area
(holes removed)
|
Table A1:6 in Praz et al, 2017 |
area_porous_r |
m^2 |
float |
Ratio area_porous
over area
|
Table A1:7 in Praz et al, 2017 |
Elliptical approximation: fitted ellipse |
||||
ell_fit_A |
m |
float |
Major dimension |
Table A1:8 in Praz et al, 2017 |
ell_fit_B |
m |
float |
Minor dimension |
Table A1:9 in Praz et al, 2017 |
ell_fit_area |
m^2 |
float |
Area |
Table A1:10 in Praz et al, 2017 |
ell_fit_ori |
deg |
float |
Orientation |
Table A1:11 in Praz et al, 2017 |
ell_fit_a_r |
float |
Axis ratio (ell_fit_A/ell_fit_B) |
Table A1:12 in Praz et al, 2017 |
|
ell_fit_ecc |
float |
Eccentricity |
Table A1:13 in Praz et al, 2017 |
|
compactness |
float |
area / ell_fit_area |
Table A1:14 in Praz et al, 2017 |
|
Elliptical approximation: inscribed ellipse (same center and orientation of fitted one) |
||||
ell_in_A |
m |
float |
Major dimension |
Table A1:15 in Praz et al, 2017 |
ell_in_B |
m |
float |
Minor dimension |
Table A1:16 in Praz et al, 2017 |
ell_in_area |
m^2 |
float |
Area |
Table A1:17 in Praz et al, 2017 |
Elliptical approximation: circumscribed ellipse (same center and orientation of fitted one) |
||||
ell_out_A |
m |
float |
Major dimension |
Table A1:18 in Praz et al, 2017 |
ell_out_B |
m |
float |
Minor dimension |
Table A1:19 in Praz et al, 2017 |
ell_out_area |
m^2 |
float |
Area |
Table A1:20 in Praz et al, 2017 |
Particle shape |
||||
roundness |
float |
Area /
circum. circle area
|
Table A1:30 in Praz et al, 2017 |
|
p_circ_out_r |
float |
Circ. circle perim/
perimeter
|
Table A1:31 in Praz et al, 2017 |
|
rectangularity |
float |
Area /
bounding box area
|
Table A1:32 in Praz et al, 2017 |
|
bbox_width |
m |
float |
Bounding box width |
Table A1:33 in Praz et al, 2017 |
bbox_len |
m |
float |
Bounding box height |
Table A1:34 in Praz et al, 2017 |
rect_perim_ratio |
float |
Bounding box perim/
perimeter
|
Table A1:35 in Praz et al, 2017 |
|
rect_aspect_ratio |
float |
|
Table A1:36 in Praz et al, 2017 |
|
rect_eccentricity |
float |
|
Table A1:37 in Praz et al, 2017 |
|
solidity |
float |
Area/convex hull area |
Table A1:38 in Praz et al, 2017 |
|
convexity |
float |
Convex hull perim.
/ perimeter ratio
|
Table A1:39 in Praz et al, 2017 |
|
hull_n_angles |
int |
# vertices con. hull |
Table A1:40 in Praz et al, 2017 |
|
p_circ_r |
float |
Perimeter/ eq. area
circle perimeter
|
Table A1:41 in Praz et al, 2017 |
|
frac_dim_boxcounting |
float |
Fractal dim. (box) |
Table A1:42 in Praz et al, 2017 |
|
frac_dim_theoretical |
float |
Fractal dim. |
Table A1:43 in Praz et al, 2017
Grazioli et al, 2014
|
|
nb_holes |
int |
# holes inside ROI |
||
Morphological skeleton |
||||
skel_N_ends |
int |
Skeleton # ending
points
|
Table A1:44 in Praz et al, 2017 |
|
skel_N_junc |
int |
Skeleton # junctions |
Table A1:45 in Praz et al, 2017 |
|
skel_perim_ratio |
float |
Skeleton length /
perimeter
|
Table A1:46 in Praz et al, 2017 |
|
skel_area_ratio |
1/m |
float |
Skeleton length
area
|
Table A1:47 in Praz et al, 2017 |
Rotational symmetry |
||||
sym_P1 … sym_P6 |
float |
Standardized dist.
to centroid Fourier
power spec. P1…P6
|
Table A1:49-54 in Praz et al, 2017 |
|
sym_Pmax_id |
int |
ID of max. (sym_P*) |
Table A1:55 in Praz et al, 2017 |
|
sym_P6_max_ratio |
float |
sym_P6 / max(sym_P*) |
Table A1:56 in Praz et al, 2017 |
|
sym_mean |
pix |
float |
Mean dist. to centroid |
Table A1:57 in Praz et al, 2017 |
sym_std |
pix |
float |
Standard deviation
dist. to centroid
|
Table A1:58 in Praz et al, 2017 |
sym_std_mean_ratio |
float |
sym_std / sym_mean |
Table A1:59 in Praz et al, 2017 |
|
Texture operators (for ROI/particle) |
||||
intensity_mean |
float |
Mean pixel brightness |
Table A1:60 in Praz et al, 2017 |
|
intensity_max |
float |
Max. pixel brightness |
Table A1:61 in Praz et al, 2017 |
|
contrast |
float |
Contrast |
Table A1:62 in Praz et al, 2017 |
|
intensity_std |
float |
std. pixel brightness |
Table A1:63 in Praz et al, 2017 |
|
hist_entropy |
float |
Bright. hist. entropy |
Table A1:64 in Praz et al, 2017 |
|
local_std |
float |
Average local (3x3)
intensity std
|
Table A1:65 in Praz et al, 2017 |
|
local_intens |
float |
Average local (3x3)
mean intensity
|
Table A1:66 in Praz et al, 2017 |
|
lap_energy |
float |
Energy of Laplacian |
Table A1:67 in Praz et al, 2017 |
|
wavs |
float |
Sum of wavelet coeff. |
Table A1:68 in Praz et al, 2017 |
|
complexity |
float |
Particle complexity |
Table A1:69 in Praz et al, 2017
Garrett and Yuter, 2014
|
|
Haralick features |
||||
har_energy |
float |
Haralick energy |
Table A1:70 in Praz et al, 2017 |
|
har_contrast |
float |
Haralick contrast |
Table A1:71 in Praz et al, 2017 |
|
har_corr |
float |
Haralick correlation |
Table A1:72 in Praz et al, 2017 |
|
har_hom |
float |
Haralick homogeneity |
Table A1:73 in Praz et al, 2017 |
|
Riming estimation information |
||||
riming_deg_level |
float |
Continuous riming
degree level
|
Rc in Praz et al, 2017 |
|
riming_class_id |
int |
Discrete riming
degree class ID
|
Praz et al, 2017
0: undefined, 1: unrimed, 2: rimed
3: densely-rimed, 4: graupel-like, 5:graupel
|
|
riming_class_prob |
float |
Riming classif
probability
|
Praz et al, 2017 |
|
riming_class_name |
string |
Discrete riming
degree class name
|
See riming_class_id |
|
Melting estimation information |
||||
melting_class_id |
int |
Discrete melting
class ID
|
Praz et al, 2017
0: dry, 1: melting
|
|
melting_prob |
float |
Melting probability |
Praz et al, 2017
If rounded, it yields melting_class_id
|
|
melting_class_name |
string |
Discrete melting class name |
See melting_class_id |
|
Hydrometeor type estimation information |
||||
snowflake_class_name |
string |
Hydrometeor class name |
Praz et al, 2017
See snowflake_class_id
|
|
snowflake_class_id |
int |
Hydrometeor
class id
|
Praz et al, 2017
1: small_particle, 2: columnar_crystal,
3: planar_crystal, 4: aggregate,
5: graupel, 6: columnar_planar_combination
|
|
snowflake_class_prob |
float |
Classification
probability
|
||
Human label information |
||||
hl_snowflake |
int |
Human-label
hydrometeor set
|
Boolean flag. If set to 1, this particle
(in this CAM view) was part of the human
labelled training set of Praz et al, 2017
for hydrometeor classification
|
|
hl_snowflake_class_id |
int |
Human-label
hydrometeor class
|
Human-labeled snowflake_class_id
for the training set of Praz et al, 2017
|
|
hl_melting |
int |
Human-label melting set |
Boolean flag. If set to 1, this particle
(in this CAM view) was part of the human
labelled training set of Praz et al, 2017
for melting identification
|
|
hl_melting_class_id |
int |
Human-label
melting class
|
Human-labeled melting_class_id
for the training set of Praz et al, 2017
|
|
hl_riming |
int |
Human-label riming |
Boolean flag. If set to 1, this particl
(in this CAM view) was part of the human
labelled training set of Praz et al, 2017
for riming degree estimation
|
|
hl_riming_class_id |
int |
Human-label
riming class
|
Human-labeled riming_class_id
for the training set of Praz et al, 2017
|
|
References#
Baker and Lawson, 2006: Baker, B. & Lawson, R. P. Improvement in Determination of Ice Water Content from Two-Dimensional Particle Imagery. Part I: Image-to-Mass Relationships.J. Appl. Meteorol. Climatol.45, 1282-1290, https://doi.org/10.1175/JAM2398.1, 2006.
Praz et al, 2017: Praz, C., Roulet, Y.-A., and Berne, A.: Solid hydrometeor classification and riming degree estimation from pictures collected with a Multi-Angle Snowflake Camera, Atmos. Meas. Tech., 10, 1335-1357, https://doi.org/10.5194/amt-10-1335-2017, 2017.
Grazioli et al, 2014: Grazioli, J., Tuia, D., Monhart, S., Schneebeli, M., Raupach, T., and Berne, A.: Hydrometeor classification from two-dimensional video disdrometer data, Atmos. Meas. Tech., 7, 2869-2882, https://doi.org/10.5194/amt-7-2869-2014, 2014.
Garrett and Yuter, 2014: Garrett, T. J. & Yuter, S. E. Observed influence of riming, temperature, and turbulence on the fallspeed of solid precipitation. Geophys. Res. Lett.41, 6515-6522, https://doi.org/10.1002/2014GL061016, 2014.