Skip to main content

Generate random Google Street View images from all around the world

Project description

Street View Randomizer

GitHub GitHub repo size

A Python command-line interface designed to generate random images from Google Street View.

Generate random Google Street View images from all around the world.

Requirements

Install

pip install street-view-randomizer

Usage

The basic usage defaults to generating a single image anywhere in the world (equal probabilities with respect to countries sizes):

street-view-randomizer --api-key=yourapikeyhere

The output will be something like:

Searched image in PRI | lon:   -65.99685876172165 lat:    18.08766139664086 | elapsed time:   399.14ms

> Image found in PRI (Puerto Rico) | lon: -66.0217481, lat: 18.1029857 | attempts: 1 | total elapsed time: 0.40s
	(1/1)	Saving to ./images/pri/-66.0217481_18.1029857_h0_p0_f90.jpg...

API key

The script will fallback to the Google Maps Platform API key in the GOOGLE_MAPS_API_KEY environment variable if the --api-key argument is not provided. You can set it in your .bashrc or .zshrc file:

export GOOGLE_MAPS_API_KEY=yourapikeyhere

General options

Here is a list of some useful flags one may pass to customize the behavior of the script.

-k, --api-key

Google Maps Platform API key.

-c, --countries

Use the -c argument together with a list of one or more ISO3 country codes to narrow the search. For instance, if we are interested in fetching an image from either Brazil, Argentina or Chile:

street-view-randomizer -c BRA ARG CHL

-l, --list-countries

Display a list of all available countries (those with some Google Street View Coverage).

-r, --radius

Defines a radius in meters centered on a latitude and longitude. The default value is 5.000 (5km). This value should only be increased if searching for an image is taking too long.

-a, --use-area

If the size of the country matters when sampling from a group of countries, passing in the -a flag will give bigger countries more chances of being drawn. The following chart shows the odds for each country if we consider the full space search:

areas_percentage

-n, --samples

To sample more than once (this doesn't mean fetching more than one image per country), pass in the -n flag with some desired number, e.g.:

street-view-randomizer -n 3

Note that the maximum number of iterations allowed is 28.000, which happens to be the maximum number of requests per month one can make without being charged by the Google Maps Platform. Be careful!

-o, --output-dir

By default all images are saved under the images directory from where the script is executed. To change the output directory, pass in the -o flag with the desired path, e.g.:

street-view-randomizer -o /home/user/images

Image options

Images will be saved to a directory named after the country code (ISO3) where the following naming convention applies:

<lon>_<lat>_h<heading>_p<pitch>_f<fov>.jpg

Please refer to the Street View Static API documentation to understand the meaning of heading, pitch and fov.

Anyway, you are allowed to pass a list of each one of these parameters to generate different imagery from the same coordinate.

-H, --headings

List of headings, e.g., -H 0 90 180 270. The default value is 0.

-P, --pitches

List of pitches, e.g., -P -35 0 35. The default value is 0.

-F, --fovs

List of fovs, e.g., -F 60 90 120. The default value is 90.

Note that the total number of images will be the product of the length of each list. For each heading, the algorithm will output an image for each pair of pitch and fov.

-S, --size

Size of the output image, defaults to 256x256. The maximum size allowed is 640x640. Each dimension must have at least a hundred pixels.

Putting it all together

The following command will perform 3 weighted samplings of 12 images of size 512x512:

street-view-randomizer -n 3 -a -H 0 90 180 270 -P -45 0 35 -S '512x512'

Contributing

The code should be pretty straightforward to understand and modify. Feel free to open an issue or submit a pull request.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

street-view-randomizer-0.1.0.tar.gz (3.5 MB view hashes)

Uploaded Source

Built Distribution

street_view_randomizer-0.1.0-py3-none-any.whl (11.0 kB view hashes)

Uploaded Python 3

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page