SidewalkMask

Masks of sidewalks are generated for google map satellite images, based on overlapping the sidewalk shapefile on top.

Getting Started

These instructions will get you a copy of the project up and running on your local machine for development and testing purposes.

Prerequisites

Install python. The code is in python3.x. If you are using python2.x, you will need a 3to2 package to convert the code into python2 syntax. Download and install 3to2 package. Run with -w flag to convert and backup original files(.bak).

sudo python setup.py build
sudo python setup.py install
3to2 files -w path/to/your/filename.py

It's also recommended to install:

QGIS Desktop.
Virtualenv for python.

Installing

Start Virtualenv

It's recommended (optional) to run your program in virtualenv:

virtualenv -p /usr/bin/python3 py3env-sidewalk
source py3env-sidewalk/bin/activate

Install python packages

Browse to SidewalkMask directory, install all packages with pip:

 $ sudo apt-get install python3-pip
 $ pip3 install -w requirements.txt

Congratulations you are ready to go!! If you runs into error, fix the error and rerun the command above. Example errors: 1.Missing modules.

 $ pip3 install psycopg2
 $ pip3 install typing 
 $ pip3 install Cython

2.Uable to install rtree in lack of libspatialindex. For libspatialindex package, download and make install. run sudo ldconfig. Then install rtree either with requirements.txt or from .whl. 3.Error: /usr/bin/ld: cannot find -lopenblas The reason is missing library: sudo apt-get install libopenblas-dev 4.Last tip: If you have error installing one package, maybe try to remove it from the requirements.txt and install the rest first.

Create Postgres database

We use postgres and postgis as the database.

$ sudo apt-get update
$ sudo apt-get install postgresql postgresql-contrib

Now test if postgres exist: ls -l /etc/init.d/post* which shows postgresql Now you can start the server sudo service postgresql start Note, if ls command returns postgres, then start server using sudo service postgres start It depends on what name your machine refers to the server. Next, install postgis.

$ sudo apt-get install -y postgis postgresql-9.3-postgis-2.11

Log into the initial account with default user: sudo -i -u postgres. Then,

$ createdb sidewalk_dataset
$ psql sidewalk_dataset
# CREATE USER sidewalk WITH PASSWORD 'sidewalk';
# GRANT ALL PRIVILEGES ON DATABASE sidewalk_dataset TO sidewalk;
# \q

Now create postgis extension for your database.

sudo -u postgres psql -c "CREATE EXTENSION postgis; CREATE EXTENSION postgis_topology;" sidewalk_dataset

Create Schema in sidewalk_dataset

Connect to sidewalk_dataset with psql sidewalk_dataset.

sidewalk_dataset=# CREATE SCHEMA sidewalk;
sidewalk_dataset=# ALTER SCHEMA sidewalk OWNER TO sidewalk;
sidewalk_dataset=# SET search_path TO <yourschema>, public, pg_catalog, topology;

If the spatial reference system is not listed in public.spatial_ref_sys, write them in a csv file and import :

sidewalk_dataset=#  COPY spatial_ref_sys FROM '/home/pothole/Downloads/spatial_ref_sys.csv' DELIMITER ',' CSV HEADER;

Export shapefile into .sql

First, decide SRID using this website by uploading the .prj file. For GCS_WGS_1984, the SRID is 4326. Now convert your shapefile into .sql file.

$ shp2pgsql -s <SRID> <shapefile> <schema>.<tablename> <db_name> > filename.sql
$ sudo -u sidewalk psql -d sidewalk_dataset -a -f filename.sql

If succeeded, you should see the tablename in your database.

sidewalk_dataset=# \dt
sidewalk_dataset=# \dn
sidewalk_dataset=# \l

Running the code

Create your own API KEY to access the google satellite images here. Paste in to the script.

$ python3 src/download_google_map.py
$ python3 src/GoogleStaticMapMasks.py

You will find the output under data/output folder.

kotarohara/SidewalkMask