Skip to content
/ hipp Public
forked from friedrichknuth/hipp

Historical Image Pre-Processing

License

Notifications You must be signed in to change notification settings

uw-cryo/hipp

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Historical Image Pre-Processing

Library to pre-process scanned historical images for Structure from Motion (SfM) surface reconstruction and photogrammetric analysis. DOI

Features

Data Query and Download

  • Download imagery from historical image archives
  • Supported archives:
    • North American Glacier Photography (NAGAP)
    • USGS Earth Explorer

Fiducial Marker Detection

  • Built-in application to create fiducial marker templates
  • OpenCV template matching to detect fiducial marker coordinates
  • Fiducial markers are detected at sub-pixel precision
  • Can detect 4 midside and/or 4 corner fiducials
  • Replaces poor matches with np.nan based on threshold
  • Computes estimated principal point
  • Quality Control
    • Outputs window image around detected fiducial marker for visual verification
    • Creates qc plots for fiducial coordinates and intersection angles before and after affine transformation

Fiducial Marker Proxy Detection

  • Routines to detect proxy for midside fiducial markers, when actual fiducial markers are cropped out of image frame

Image Restitution

  • Computes affine transform between calibrated (true) fiducial marker coordinates and detected coordinates
  • Affine transforms images
    • Requires minimum of 3 successfully detected fiducial markers to perform restitution
  • Crops images about principal point to standard size
  • Contrast Limited Adaptive Historgram Equalization (CLAHE) to improve match point detection during SfM processing

Examples

See notebooks for processing examples.

Installation

$ git clone https://github.com/friedrichknuth/hipp.git
$ cd ./hipp
$ conda env create -f environment.yml
$ conda activate hipp
$ pip install -e .

References

Bradski, G. (2000). "The OpenCV Library". Dr. Dobb's Journal of Software Tools.

About

Historical Image Pre-Processing

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 100.0%