mmaster_tools

pymmaster.mmaster_tools provides most of the routines used for removing jitter-related bias in ASTER DEMs produced using MicMac ASTER.

pymmaster.mmaster_tools.calculate_dH(mst_dem, slv_dem, pts)

Calculate difference between master and slave DEM. Master and slave DEMs must have the same shape. First, differences are calculated, then data are filtered using a 5x5 median filter, and differences greater than 100 m, or differences on high (>50deg) or low (< 0.5deg) slopes are removed.

Parameters:

• mst_dem – master DEM (i.e., external DEM/ICESat data)
• slv_dem – slave DEM (i.e., ASTER scene)
• pts – True if master DEM is ICESat data, False if master DEM is a DEM.

Returns:

dH: filtered elevation differences between master and slave DEMs.

pymmaster.mmaster_tools.fitfun_sumofsin(xx, p)

The Base function for the sum of sines fitting to elevation differences in the along track (xx) direction of the ASTER flightpath. DEPRECATED - uses only one angle instead of two. see fitfun_sumofsin_2angle

pymmaster.mmaster_tools.fitfun_sumofsin_2angle(xxn, xxb, p)

Fit sum of sines function for two track angles.

Parameters:

• xxn (array-like) – along-track distance in nadir direction
• xxb – along-track distance in back-looking direction
• p (array-like) – sum of sines parameters

Returns sum_of_sines:  

sum of sines evaluated at the given along-track distances.

pymmaster.mmaster_tools.get_aster_footprint(gran_name, proj4='+units=m +init=epsg:4326', indir=None, polyout=True)

Create shapefile of ASTER footprint from .met file

Parameters:

• gran_name (str) – ASTER granule name to use; assumed to also be the folder in which .met file(s) are stored.
• proj4 (str, dict) – proj4 representation for coordinate system to use [default: ‘+units=m +init=epsg:4326’, WGS84 Lat/Lon].
• indir (str) – Directory to search in [default: current working directory].
• polyout (bool) – Create a shapefile of the footprint in the input directory [True].

Returns footprint:  

shapely Polygon representing ASTER footprint, reprojected to given CRS.

pymmaster.mmaster_tools.get_atrack_coord(mst_dem, myangN, myangB)

Createx numpy arrays for along- and cross-track distances from DEM and angle maps.

Parameters:

• dem (pybob GeoImg) – DEM to get x,y positions from
• myangN (pybob GeoImg) – MMASTER 3N track angle (i.e., track angle in nadir image)
• myangB (pybob GeoImg) – MMASTER 3B track angle (i.e., track angle in back-looking image)

Returns yyn, yyb:  

along-track distances in 3N and 3B track directions.

pymmaster.mmaster_tools.get_fit_variables(mst_dem, slv_dem, xxn, pts, xxb=None)

Get input variables for bias fitting.

Parameters:

• mst_dem (pybob GeoImg, ICESat) – Master DEM to use for fitting
• slv_dem (pybob GeoImg) – Slave DEM to use for fitting
• xxn (array-like) – along- or cross-track distance in the nadir track direction
• pts (bool) – True if mst_dem is ICESat points, False if mst_dem is a GeoImg.
• xxb (array-like) – along- or cross-track direction in the back-looking track direction. If not provided, fit variables are provided only for one angle.

Returns xx, dH, grp_xx, grp_dH, xx2, grp_sts:  

track distance, elevation differences, group distances, group median dH group values, group distances in the back-looking direction, group statistics.

pymmaster.mmaster_tools.get_group_statistics(invar, indata, indist=500)

Calculate statistics of groups of pixels grouped by along-track distance.

Parameters:

• invar (array-like) – Input array that determines how to group data (i.e., along-track distance)
• indata (array-like) – Data to calculate group statistics for.
• indist (float) – Distance by which to group pixels.

Returns grp_stats:  

group statistics for input data.

pymmaster.mmaster_tools.get_track_angle(fprint, track_dist)

Calculate the angle made by the ASTER flight track from scene footprint.

Parameters:

• fprint (shapely Polygon) – Footprint of ASTER scene
• track_dist (float) – Distance along flight track within scene at which to calculate angle.

Returns track_angle:  

angle, in degrees, of flight track

pymmaster.mmaster_tools.get_xy_rot(dem, myang)

Rotate x, y axes of image to get along- and cross-track distances.

Parameters:

• dem (pybob GeoImg) – DEM to get x,y positions from.
• myang (float) – angle by which to rotate axes

Returns xxr, yyr:  

arrays corresponding to along (x) and cross (y) track distances.

pymmaster.mmaster_tools.make_group_id(inmat, grpid)

Make a unique ID for statistical analysis.

Parameters:

• inmat (array-like) – Input array to create ID for.
• grpid (array-like) – Array of input group IDs

Returns outmat:

Output of group IDs

pymmaster.mmaster_tools.make_mask(inpoly, pts, raster_out=False)

Create a True/False mask to determine whether input points are within a polygon.

Parameters:

• inpoly (shapely Polygon) – Input polygon to use to create mask.
• pts (array-like, pybob.GeoImg) – Either a list of (x,y) coordinates, or a GeoImg. If a list of (x,y) coordinates, make_mask uses shapely.polygon.contains to determine if the point is within inpoly. If pts is a GeoImg, make_mask uses ogr and gdal to “rasterize” inpoly to the GeoImg raster.
• raster_out (bool) – Kind of output to return. If True, pts must be a GeoImg.

Returns mask:

An array which is True where pts lie within inpoly and False elsewhere.

pymmaster.mmaster_tools.mask_raster_threshold(rasname, maskname, outfilename, threshold=50, datatype=<MagicMock id='139924155742248'>)

Filter values in an input raster if the correlation mask raster falls below a threshold value. Removes noise from low-correlation areas (i.e., clouds) using a binary opening operation.

Parameters:

• rasname (str, pybob.GeoImg) – path to raster (i.e., DEM) to mask, or GeoImg object.
• maskname (str, pybob.GeoImg) – path to correlation raster to determine mask values, or GeoImg object.
• outfilename (str) – filename to write masked raster to.
• threshold (float) – correlation value to determine masked values.
• datatype (numpy datatype) – datatype to save raster as.

Returns masked_ras:  

GeoImg of masked input raster.

pymmaster.mmaster_tools.mmaster_bias_removal(mst_dem, slv_dem, glacmask=None, landmask=None, pts=False, work_dir='.', tmp_dir=None, out_dir=None, return_geoimg=True, write_log=False, zipped=False, robust=True)

Removes cross track and along track biases from MMASTER DEMs.

Parameters:

• mst_dem (str, pybob.GeoImg, pybob.ICESat) – Path to filename or GeoImg dataset representing “master” DEM or ICESat.
• slv_dem (str, pybob.GeoImg) – Path to filename or GeoImg dataset representing “slave” DEM (developed for ASTER).
• glacmask (str) – Path to shapefile representing points to exclude from co-registration consideration (i.e., glaciers).
• landmask (str) – Path to shapefile representing points to include in co-registration consideration (i.e., stable ground/land).
• pts (bool) – If True, program assumes that masterDEM represents point data (i.e., ICESat), as opposed to raster data. Slope/aspect are then calculated from slaveDEM. masterDEM should be a string representing an HDF5 file continaing ICESat data.
• work_dir (str) – Location where output files and directory should be saved [.]
• tmp_dir – Location where files are processed
• out_dir (str) – Location to save bias removal outputs.
• return_geoimg (bool) – Return GeoImg objects of the corrected slave DEM and the co-registered master DEM [True]
• write_log (bool) – Re-direct stdout, stderr to a log file in the work directory [False]
• zipped (bool) – extract from zip archive, keep a minimum of output files (logs and pdfs only) [False]
• robust (bool) – solve low-frequency and jitter components of along-track bias separately. If spatial sampling is limited, this is a better approach than solving along-track as one single step.

Returns slv_corr, mst_coreg:  

corrected MMASTER DEM, co-registered master DEM (if return_geoimg)

pymmaster.mmaster_tools.nmad(data)

Calculate the Normalized Median Absolute Deviation (NMAD) of the input dataset.

Parameters:data (array-like) – input data on which to calculate NMAD Returns nmad:NMAD of input dataset.

pymmaster.mmaster_tools.orient_footprint(fprint)

Orient ASTER footprint coordinates to be clockwise, with upper left coordinate first.

Parameters:fprint (shapely polygon) – footprint to orient Returns o_fprint:  re-oriented copy of input footprint.

pymmaster.mmaster_tools.plot_bias(xx, dH, grp_xx, grp_dH, title, pp, pmod=None, smod=None, plotmin=None, txt=None)

data : original data as numpy array (:,2), x = 1col,y = 2col gdata : grouped data as numpy array (:,2) pmod,smod are two model options to plot, numpy array (:,1)

pymmaster.mmaster_tools.polynomial_fit(x, y)

[DEPRECATED] A polynomial search function for orders 1-6 given dependent (x) and independent (y) variables. Uses the numpy polynomial package.

pymmaster.mmaster_tools.preprocess(mst_dem, slv_dem, glacmask=None, landmask=None, work_dir='.', out_dir='biasrem', pts=False)

Pre-process ASTER scene to enable cross- and along-track corrections. Co-registers the ASTER (slave) and external (master) DEMs/ICESat, and shifts the orthoimage and correlation mask based on the offset calculated. Results are saved by default in a folder called ‘coreg’ which is moved to the ‘bias_removal’ folder at the end of this function

Parameters:

• mst_dem (str, pybob.Geoimg, pybob.ICESat) – Path to filename or GeoImg dataset representing external “master” DEM or ICESat dataset.
• slv_dem (str, pybob.GeoImg) – Path to filename or GeoImg dataset representing ASTER DEM
• glacmask (str) – Path to shapefile representing areas to exclude from co-registration consideration (i.e., glaciers).
• landmask (str) – Path to shapefile representing areas to include in co-registration consideration (i.e., stable ground/land).
• work_dir – Working directory to use [Assumes ‘.’].
• out_dir (str) – Output directory for the new coregistration folder
• pts – True if mst_dem is point data, False if mst_dem is a DEM.

Returns mst_coreg, slv_coreg, shift_params:  

co-registered master, slave datasets, as well as a tuple containing x,y,z shifts calculated during co-registration process.

pymmaster.mmaster_tools.reproject_geometry(src_data, src_crs, dst_crs)

Reproject a geometry object from one coordinate system to another.

Parameters:

• src_data (shapely geometry) – geometry object to reproject
• src_crs (str, dict) – proj4 description of source CRS
• dst_crs (str, dict) – proj4 description of destination CRS

Returns dst_data:  

reprojected data.

pymmaster.mmaster_tools.robust_polynomial_fit(xx, yy)

Given sample data xx, yy, compute a robust polynomial fit to the data. Order is chosen automatically by comparing residuals for multiple fit orders.

Parameters:

• xx (array-like) – input x data (typically across-track distance)
• yy (array-like) – input y data (typically elevation differences)

Returns coefs, order:  

polynomial coefficients and order for the best-fit polynomial