Skip to main content

Python implementation of Continuous Change Detection

Project description

PyCCD - Python Continuous Change Detection

pyccd creates change segments from Landsat timeseries data

Using PyCCD

>>> import ccd
>>> results = ccd.detect(dates, blues, greens, reds, nirs, swir1s, swir2s, thermals, qas, prev_results)
>>>
>>> type(results)
<class 'dict'>
>>>
>>> results
{algorithm: 'pyccd:x.x.x',
 processing_mask: [bool, bool, ...],
 snow_prob: float,
 water_prob: float,
 cloud_prob: float,
 change_models: [
     {start_day: int,
      end_day: int,
      break_day: int,
      observation_count: int,
      change_probability: float,
      curve_qa: int,
      blue:      {magnitude: float,
                  rmse: float,
                  coefficients: (float, float, ...),
                  intercept: float},
      green:     {magnitude: float,
                  rmse: float,
                  coefficients: (float, float, ...),
                  intercept: float},
      red:       {magnitude: float,
                  rmse: float,
                  coefficients: (float, float, ...),
                  intercept: float},
      nir:       {magnitude: float,
                  rmse: float,
                  coefficients: (float, float, ...),
                  intercept: float},
      swir1:     {magnitude: float,
                  rmse: float,
                  coefficients: (float, float, ...),
                  intercept: float},
      swir2:     {magnitude: float,
                  rmse: float,
                  coefficients: (float, float, ...),
                  intercept: float},
      thermal:   {magnitude: float,
                  rmse: float,
                  coefficients: (float, float, ...),
                  intercept: float}}
                 ]
}

Default processing parameters can be overridden using a dictionary (see parameters.yaml for valid keys):

>>> import ccd
>>> params = {'QA_BITPACKED': False,
              'QA_FILL': 255,
              'QA_CLEAR': 0,
              'QA_WATER': 1,
              'QA_SHADOW': 2,
              'QA_SNOW': 3,
              'QA_CLOUD': 4}
>>> results = ccd.detect(dates, blues, greens, reds, nirs, swir1s, swir2s, thermals, qas, params=params)

Installing

System requirements (Ubuntu)

  • python3-dev
  • gfortran
  • libopenblas-dev
  • liblapack-dev
  • graphviz
  • python-virtualenv

System requirements (Centos)

  • python3-devel
  • gfortran
  • blas-dev
  • lapack-dev
  • graphviz
  • python-virtualenv

It's highly recommended to do all your development & testing in a virtual environment.

user@dev:/home/user/$ mkdir pyccd
user@dev:/home/user/$ cd pyccd
user@dev:/home/user/pyccd$ virtualenv -p python3 .venv
user@dev:/home/user/pyccd$ . .venv/bin/activate
(.venv) user@dev:/home/user/pyccd$
Install
$ pip install -e .[test,dev,docs,deploy,profile]

Testing

$ pytest
$ pytest --profile
$ pytest --profile-svg

# pytest-watch
$ ptw

Make targets

$ make build
$ make tests
$ make docs
$ make deploy
$ make profile

Profiling

Decorate the function to be profiled with @profile and run make profile. Remove decorations before committing code.

Contributing

Before committing to this repository, run the following command.

git config --local commit.template .gitmessage

This will add the LCMAP commit template to git commit.

jira/lcmap-xxxx: Title
Description

Contributions are most welcome.

  1. Open an issue and discuss the change.
  2. Branch from major version and name it after the issue
    • jira/lcmap-xxxx
  3. Write automated tests for your changes and make sure all tests pass.
  4. Update documentation in project.
  5. Submit pull request to the appropriate branch (e.g. 1,x)

Versions

PyCCD previously followed MAJOR.MINOR.PATCH.LABEL semantic versioning but has changed to date based semantic versioning, thus: YYYY.MM.DD[.HH.MM.SS][-label].

PyCCD's version is defined by the ccd/version.py/__version__ attribute.

References

Project details


Download files

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

Files for lcmap-pyccd, version 2020.10.10
Filename, size File type Python version Upload date Hashes
Filename, size lcmap_pyccd-2020.10.10-py3-none-any.whl (30.8 kB) File type Wheel Python version py3 Upload date Hashes View
Filename, size lcmap-pyccd-2020.10.10.tar.gz (30.5 kB) File type Source Python version None Upload date Hashes View

Supported by

AWS AWS Cloud computing Datadog Datadog Monitoring DigiCert DigiCert EV certificate Facebook / Instagram Facebook / Instagram PSF Sponsor Fastly Fastly CDN Google Google Object Storage and Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Salesforce Salesforce PSF Sponsor Sentry Sentry Error logging StatusPage StatusPage Status page