.gitignore | ||
aggregate-rivers.sql | ||
amalgamate1.png | ||
bib.bib | ||
COPYING | ||
db | ||
Dockerfile | ||
extract-and-generate | ||
gdr2pgsql | ||
in-container | ||
init.sql | ||
layer2img.py | ||
Makefile | ||
mj-msc-full.pdf | ||
mj-msc.tex | ||
openmap-wm-bad.png | ||
openmap-wm-good.png | ||
README.md | ||
rivers-10.sql | ||
rivers-50.sql | ||
rivers-250.sql | ||
salvis.png | ||
slides-2021-03-29.txt | ||
test-rivers.sql | ||
test.sql | ||
vars.awk | ||
visuals.sql | ||
vu.pdf | ||
wang125.png | ||
wm.sql |
Wang–Müller line generalization algorithm in PostGIS
This is Wang–Müller line generalization algorithm implementation in PostGIS. Following "Line generalization based on analysis of shape characteristics" by the same authors, 1998.
Status
Mostly works. Read mj-msc-full.pdf
for visual examples and possible gotchas.
Structure
There are 2 main deliverables:
wm.sql
, the implementation.- paper
mj-msc-full.pdf
, a MSc thesis, explaining it.
It contains a few supporting files, notably:
tests.sql
synthetic unit tests.test-rivers.sql
tests with real rivers.Makefile
glues everything together.layer2img.py
converts a PostGIS layer to an embeddable image.aggregate-rivers.sql
combines multiple river objects (linestrings or multilinestrings) to a single one.init.sql
initializes PostGIS database for running the tests.rivers-*.sql
are national dataset snapshots of rivers (Makefile
contains code to update them).- ... and a few more files necessary to build the paper.
Running
make help
lists the select commands for humans. As of writing:
# make help
clean Clean the current working directory
clean-tables Remove tables created during unit or rivers tests
help Print this help message
mj-msc-full.pdf Thesis for publishing
mj-msc-gray.pdf Gray version, to inspect monochrome output
refresh-rivers Refresh river data from national datasets
test-rivers Rivers tests (slow)
test Unit tests (fast)
To execute the algorithm, run:
make test
for tests with synthetic data.make test-rivers
for tests with real rivers. You may adjust the rivers and data source (e.g. use a different country instead of Lithuania) by changing theMakefile
and the test files. Left as an exercise for the reader.
N.B. the make test-rivers
fails (see test-rivers.sql
), because with higher
dhalfcircle
values, the unionized river (salvis
) is going on top of itself,
making the resulting geometry invalid during the process.
Building the paper (pdf)
# make -j mj-msc-full.pdf
mj-msc.tex
results in mj-msc-full.pdf
. This step needs quite a few
or a container: see Dockerfile
for dependencies or in-container
to run
it all in the container.
Contributing
This repository does not accept contributoins. Please fork it. If a fork has improved the algorithm substantially, you are welcome to ping me, I will link to it in this README.
Credit
Nacionalinė Žemės Tarnyba for the river data sets.
License
GPLv2 or later.