Parse format: "1:focal_length,optical_center 2:focal_length,other_intrinsics 3:none" Applies when optimizing intrinsics per sensor. The numbers are se

Finish standard deviation calculation. At this point, tile contains sum of squared deviations from mean, and weights contains the count of valid pixel

Apply the fisheye distortion model. Input and output are normalized pixels.

See the .h file for the documentation

Fit a CSM camera to an optical bar camera. Optical bar cameras assumes the scan lines are vertical, as for KH-9, which differs with how a CSM model is

Find the best camera that fits the current GCP

Ensure that quaternions don't suddenly change sign. This is a bugfix.

Boost does not like a time string such as "2017-12-07 15:36:40.90795Z" because it expects precisely 6 digits after the dot. Fix that.

Fix seams by finding interest point matches and adjusting the transforms so that they map points from one image on top of matches from another image.

GDAL block write sizes must be a multiple to 16 so if the input value is not a multiple of 16 increase it until it is.

Flag outliers by triangulation angle and reprojection error. It is assumed that the cameras in world_to_cam are up-to-date given the current state of

Flag outliers by reprojection error with input cameras. This assumes that the input cameras are pretty accurate.

If input is myRun/ortho, return "ortho". Same for "fireball", "lidar", etc.

Given a command-line string, form argc and argv. In addition to pointers, will also store the strings in argv_str, to ensure permanence.

Form a file with given numbers. This is used in more than one place.

Assemble the polygon structure

For frame cameras that belong to the same orbital group, collect together the initial positions in a single vector, and same for quaternions. Linescan

Form a QImage to show on screen from a single-channel double image. Scale pixel values to [0, 255] and handle nodata. If colormap is not null, apply i

Form a QImage from a two-channel uint8 image (grayscale + mask).

Put the triangulated points in a vector. Update the cnet from the DEM, if we have one.

Create a camera using user-specified options. Keep record of the georeference and the datum.

Given a frame linescan camera and the transform from it to the current camera, find the current camera to world transform as an array.

Must all this after optimization to update the Extrinsics structure

Gamma correction for x between 0 and 1. https://en.wikipedia.org/wiki/SRGB#Specification_of_the_transformation

A function that will take as input the endpoints and will compute the positions and orientations in ECEF. The key observation is that the positions wi

Create and save a linescan camera with given camera positions and orientations. There will be just one of them, as all poses are part of the same line

A function to create and save Pinhole cameras. Assume no distortion, and pixel pitch = 1.

Generate the cameras and images for a rig

Generate a temporary directory for HDF extraction. If the output_prefix has a parent path (e.g., "run/out"), create "run/tmp". Otherwise, create a uni

Produce and write a GCP file. Can throw exceptions. TODO(oalexan1): Consider moving this to WV, together with the logic for reading GCPs. Support read

Generate a list of all the processing batches required for a run

Generates a KML plot of a point border

Generate a folder containing handy debugging files including output thumbnails

Generates a KML plot for the input points

Generate a single output tile. This creates a DemMosaicView, rasters it to disk, and removes the tile if it contains no valid pixels.

Generate the name of a tile based on its location

Generates the set of GDC/pixel pairs that will be fed into the solver.

Generates the set of GDC/pixel pairs that will be fed into the solver.

Generate lon-lat-height to image pixel correspondences that we will use to create the RPC model.

Return the affine georeferencing transform.

Find the Jacobian of geodetic_to_pixel using numerical differentiation. This is used for testing purposes.

Builds a GeoReference from the first cloud having a georeference in the list

Returns a list of the currently active jobs and their status

Return the file name used to tar up the generated bundle-adjustd and aligned camera files

Get files with given extension. This returns the files without sorting or the folder name prepended to them.

Return the email address to use for a user

Return EPSG code for a location. See notes in getProjString.

Try to figure out the reason that a DEM failed

This function simplifies the logic when we know the info we need is surely in the first child.

Obtains some image geo information from gdalinfo in dictionary format

Return a list containing all the currently stored image files

Find a good image stereo spacing interval that gives us a good balance between coverage and baseline width. Also detect all frames where

Return the file name used to tar up the downloaded input data

Returns (minLon, maxLon, minLat, maxLat) for an ISIS compatible object

Return the file name used to tar up the label files

Path to the OSSP label training file

Return the path to the latest log file matching a prefix

Return a list containing all the currently stored lidar files. This does not return converted csv files.

Extracts the remainder of a line after a prefix

Given a projected proj4 string, get the longlat proj4 string

Get the path of the match file generated by bundle_adjust for a given image pair, and the path to where that match file could be copied

Get the mean distance between the input DEMs to the main DEM.

Return the number of cores available for a given node type

Return the file name used to tar up the generated camera files

Get tif files in given directory, ignoring _sub files. This returns the files without sorting.

Find all the DEM or ORTHO files unpacked from their tarball.

Parse the output from running stereo and return the search range width and the memory usage in GB.

Read the camera height above ground from the ortho file, unless it was user-specified.

Get the label file and the camera file for a given frame

Get camera center at a given time

Get camera velocity at a given time

Returns the CCD number of a filename.

Compute the offset in m_intrinsics_vec to the requested data