spymicmac.register

spymicmac.register is a collection of tools for registering images and finding GCPs.

spymicmac.register.get_imlist(im_subset)[source]

Given either a list of filenames or a regex match pattern, return a list of filenames and a pattern to provide to MicMac.

Parameters

im_subset (list) – a list of filenames or a match pattern (e.g., [‘OIS.*tif’]) representing filenames

Returns

  • imlist (list) – the list of filenames matching the provided pattern.

  • match_pattern (str) – the match pattern to be provided to MicMac.

spymicmac.register.get_mask(footprints, img, imlist, landmask=None, glacmask=None)[source]

Create a mask for an image from different sources.

Parameters

  • footprints (GeoDataFrame) – vector data representing image footprints

  • img (GeoImg) – the GeoImg to create a mask for

  • imlist (array-like) – a list of image names

  • landmask (str) – path to file of land outlines (i.e., an inclusion mask)

  • glacmask (str) – path to file of glacier outlines (i.e., an exclusion mask)

Returns

  • mask (array-like) – the mask

  • fmask (GeoImg) – the georeferenced footprint mask

  • img (GeoImg) – the GeoImg, cropped to a 10 pixel buffer around the image footprints

spymicmac.register.get_utm_str(epsg)[source]

Given a GeoImg, read the associated EPSG code and return a UTM zone.

Examples:

  • get_utm_str(32608) -> ‘UTM Zone 8N’

  • get_utm_str(32708) -> ‘UTM Zone 8S’

Parameters

epsg (int|str) – a str or int representing an EPSG Code

Returns

  • utm_str (str) – the UTM string representation

spymicmac.register.register_ortho(fn_ortho, fn_ref, fn_reldem, fn_dem, glacmask=None, landmask=None, footprints=None, im_subset=None, block_num=None, ori='Relative', ortho_res=8.0, imgsource='DECLASSII', density=200, out_dir=None, allfree=True)[source]

Register a relative orthoimage and DEM to a reference orthorectified image and DEM.

Parameters

  • fn_ortho (str) – path to relative orthoimage

  • fn_ref (str) – path to reference orthorectified image

  • fn_reldem (str) – path to relative DEM

  • fn_dem (str) – path to reference DEM

  • glacmask (str) – path to file of glacier outlines (i.e., an exclusion mask)

  • landmask (str) – path to file of land outlines (i.e., an inclusion mask)

  • footprints (str) – path to shapefile of image outlines. If not set, will download from USGS.

  • im_subset (str) – subset of raw images to work with

  • block_num (str) – block number to use if processing multiple image blocks

  • ori (str) – name of orientation directory (after Ori-) (default: Relative)

  • ortho_res (float) – approx. ground sampling distance (pixel resolution) of ortho image (default: 8 m)

  • imgsource (str) – USGS dataset name for images (default: DECLASSII)

  • density (int) – pixel spacing to look for GCPs (default: 200)

  • out_dir (str) – output directory to save auto GCP files to (default: auto_gcps)

  • allfree (bool) – run Campari setting all parameters free (default: True)

spymicmac.register.sliding_window_filter(img_shape, pts_df, winsize, stepsize=None, mindist=2000, how='residual', is_ascending=True)[source]

Given a DataFrame of indices representing points, use a sliding window filter to keep only the ‘best’ points within a given window and separation distance.

Parameters

  • img_shape (array-like) – The shape of the image to filter indices from.

  • pts_df (DataFrame) – A DataFrame that contains the pixel locations of points (column: orig_j, row: orig_i)

  • winsize (int) – the size of the window to use for the filter.

  • stepsize (int) – how large of a step size to use for the sliding window (default: winsize/2)

  • mindist (int) – the minimum distance (in pixels) to use between points (default: 2000)

  • how (str) – how to sort pts_df to determine the ‘best’ points to keep (default: ‘residual’)

  • is_ascending (bool) – whether the column named in ‘how’ should be sorted ascending or descending (default: True)

Returns

  • out_inds (array-like) – an array with the filtered indices

spymicmac.register.transform_centers(img_gt, ref, imlist, footprints, ori)[source]

Use the camera centers in relative space provided by MicMac Orientation files, along with camera footprints, to estimate a transformation between the relative coordinate system and the absolute coordinate system.

Parameters

  • img_gt (array-like) – the image GeoTransform (as read from a TFW file)

  • ref (GeoImg) – the reference image to use to determine the output image shape

  • imlist (list) – a list of of the images that were used for the relative orthophoto

  • footprints (GeoDataFrame) – the (approximate) image footprints - the centroid will be used for the absolute camera positions.

  • ori (str) – name of orientation directory

Returns

  • model (AffineTransform) – the estimated Affine Transformation between relative and absolute space

  • inliers (array-like) – a list of the inliers returned by skimage.measure.ransac

  • join (GeoDataFrame) – the joined image footprints and relative orientation files