Skip to main content

A python package for landslides research

Project description

Introduction

Landsifier is a Python based library to estimate likely triggers of mapped landslides. The Beta version of library consitute three machine learning based method for finding the trigger of Landslide inventories.

  • Geometric feature based method
  • Topological feature based method
  • Image based method

Sample output of each lmethod

The below plot shows the probability of each landslide polygons in testing inventory belonging to earthquake and rainfall-induced class. The majority trigger of landslides is the final trigger of the testing inventory.

(1) Geometric feature based method

This method is based on using 2D landslide polygon geometric properties for classification. This method calculates various geometric properties of landslide polygon and these geometric properties are used as a feature space for machine learning based algorithm.

Sample landslide polygons

The below plot shows the sample landslide polygons of earthquake and rainfall-induced inventories.

Geometric properties of landslide polygon

The geometric properties of landslide polygons used are:-

  • Area (A) of landslide Polygon
  • Perimetre (P) of Landslide Polygon
  • Ratio of Area (A) to Perimetre(P)
  • Convex hull based measures (Ratio of area of polygon to area of convex hull fitted to polygon)
  • Width of minimum area bounding box fitted to polygon
  • Eccentricity of ellipse fitted to polygon having area A and perimetre P
  • minor-axis of ellipse fitted to polygon having area A and perimetre P

The below plot shows the various geometric properties of landslide polygon

(2) Topological feature based method

This method convert 2D landslide polygon to 3D landslide shape by including elevation information.Landsifier library compute topological features of 3D landslide shape use topological data analysis. These topological features can be used in machine learning algorithm for landslide triggers classification.

Sample 3D landslide shape

The below plot shows the sample 3D landslide shape of earthquake and rainfall-induced inventories.

Coversion of 2D landslide polygons to 3D landslide shape

The below plot shows the method for coversion of 2d landslide polygon to 3D shape.

(3) Image based method

This method convert landslide polygon data to landslide polygon Images. These converted landslide images are used as a input to Convolutional Neural Networks for landslide classification.

Sample Polygon Images

The below plot shows the sample landslide polygon Images.

Coversion of 2D landslide polygons to Images

The below plot shows the method for coversion of 2d landslide polygon to grayscale binary images.

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

landsifier-1.0.0.tar.gz (12.9 kB view details)

Uploaded Source

Built Distribution

landsifier-1.0.0-py3-none-any.whl (16.4 kB view details)

Uploaded Python 3

File details

Details for the file landsifier-1.0.0.tar.gz.

File metadata

  • Download URL: landsifier-1.0.0.tar.gz
  • Upload date:
  • Size: 12.9 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/4.0.1 CPython/3.9.12

File hashes

Hashes for landsifier-1.0.0.tar.gz
Algorithm Hash digest
SHA256 1b640c1de5f2c7750f04d0426b9edd0880178720bf9eb35725cbadf14930e828
MD5 ec1d656271ebb2779483ee3cc3e4b5f0
BLAKE2b-256 d47d3a3893beea69f88642a56a76d34ebf2876c2a63c766a3269a0999c97b859

See more details on using hashes here.

File details

Details for the file landsifier-1.0.0-py3-none-any.whl.

File metadata

  • Download URL: landsifier-1.0.0-py3-none-any.whl
  • Upload date:
  • Size: 16.4 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/4.0.1 CPython/3.9.12

File hashes

Hashes for landsifier-1.0.0-py3-none-any.whl
Algorithm Hash digest
SHA256 7ddb6476e0cbd611aa2ea6ee232b7b85668d86732eae9a37547cabc07b0d2144
MD5 88f9e35847cca8933048a183db597b9f
BLAKE2b-256 469da9c89d1a66e563047b0074ac249df32f7ea85bebecb55db1788582dd6ba3

See more details on using hashes here.

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