Python Client API for Web Time Series Service

WTSS is a lightweight web service for handling remote sensing imagery as time series. Given a location and a time interval you can retrieve the according time series as a Python list of real values.

If you want to know more about WTSS service, visit the Earth Observation Web Services homepage and the WTSS specification.

There are also client APIs for other programming languages: R, JavaScript, and C++.

Installing wtss.py

Please, open a shell script and try:

sudo pip install wtss

or

sudo easy_install wtss

Building and installing wtss.py from source

1. Open a shell script and go to the folder src.

2. In the shell, type:

$ sudo pip install .

That's it!

Using wtss.py to retrieve the time series

Import the wtss class and then use it to create an objet to retrieve the time series as shown in the following example:

from wtss import wtss

w = wtss("http://www.dpi.inpe.br/tws")

cv_list = w.list_coverages()

print(cv_list)

cv_scheme = w.describe_coverage("mod13q1_512")

print(cv_scheme)

ts = w.time_series("mod13q1_512", ("red", "nir"), -12.0, -54.0, "", "")

print(ts["red"])

print(ts["nir"])

print(ts.timeline)

If you want to plot a time series, you can write a code like:

import matplotlib.pyplot as pyplot
import matplotlib.dates as mdates
from wtss import wtss

w = wtss("http://www.dpi.inpe.br/tws")

# retrieve the time series for location with longitude = -54, latitude =  -12
ts = w.time_series("mod13q1_512", "red", -12.0, -54.0, start_date="2001-01-01", end_date="2001-12-31")

fig, ax = pyplot.subplots()

ax.plot(ts.timeline, ts["red"], 'o-')

fig.autofmt_xdate()

pyplot.show()

The codesnippet above will result in a chart such as:

More examples can be found in the examples directory.

References

VINHAS, L.; QUEIROZ, G. R.; FERREIRA, K. R.; CÂMARA, G. Web Services for Big Earth Observation Data. In: BRAZILIAN SYMPOSIUM ON GEOINFORMATICS, 17. (GEOINFO), 2016, Campos do Jordão, SP. Proceedings... 2016.