gipfel - photogrammetry for mountain images

Screenshot

The screenshot shows the view from Olympiaberg in Munich towards the alps. The GPS data comes from Jonathan de Ferranti's great website http://www.viewfinderpanoramas.org/.

Description

gipfel helps to find the names of mountains or points of interest on a picture.
It uses a database containing names and GPS data. With the given viewpoint (the point from which the picture was taken) and two known mountains on the picture, gipfel can compute all parameters needed to compute the positions of other mountains on the picture.
gipfel can also be used to play around with the parameters manually. But it's pretty difficult to find the right parameters for a given picture manually.
You can think of gipfel as a georeferencing software for arbitrary images (not only satellite images or maps).
gipfel also has an image stitching mode, which allows to generate panorama images from multiple images that have been referenced with gipfel. The nice thing about the new image stitching is that it reuses the code that already existed for referencing images.

Download

Source: gipfel-0.5.5.tar.gz
sha256 2cd560faaeaa310abf951ba700c2f55d1727793ed630c4027f04a067358094a4

pkgsrc:
A pkgsrc package is also available.

Source Repository

A git repository can be found here.

Requirements

• cargo
• gipfel works on UNIX-like systems (e.g. Linux, *BSD and probably others)

Changes from 0.4.1 to 0.5.5

• rust based rewrite of gipfel using libcosmic with the help of local AI (qwen3.8-flash-next)
• zooming of images
• replacement of the direction dial with a compass ruler under the image
• replacement of the GSL dependency with a own LSQ implementation
• support for GPS tracks in GPX format
• new gipfel-cli binary for stitching operations

Video

Here is an video (15MB) that demonstrates how gipfel works.

A picture taken from Olympiaberg in Munich towards the alps is used.

• after opening the image in gipfel the position is taken automatically from the Exif data
• then I increase the visibility and point it roughly to the south such that the first peak markers appear
• I then open the find peak dialog by righ clicking and look for a peak I already recognize
• after that is positioned I check for another known peak and position it
• then things look already pretty good and I just do some more adjustments for a better fit

Panorama Images

The following images are stitched together with the gipfel stitching option.

for i in *.jpg; do gipfel-cli stitch -b -t tiff/$i $i; done

was used to create correctly oriented tiff images.
These tiff images were then blended using enblend.

The following one is from Sentinel Dome (Yosemite) with USGS data: