Skip to content
master
Switch branches/tags
Code

Latest commit

 

Git stats

Files

Permalink
Failed to load latest commit information.
Type
Name
Latest commit message
Commit time
bin
 
 
src
 
 
 
 
 
 

Algorithmia vegetation detection

Project Status: WIP – Initial development is in progress, but there has not yet been a stable, usable release suitable for the public.

Part of trees/ungp

This algorithm accepts as input a directory of images hosted on algorithmia, amazon s3 or dropbox and returns the percentage vegetation detected in each image.

Running

With algorithmia algo client:

algo run nocturne/vegetation/$(git rev-parse HEAD) -d \
  '{"src": "data://.my/test_images", "method": "deep"}'

The algo. accepts 2 arguments:

  • src: The location of your street-level images.
  • method: The segmentation method to use.

Segmentaton methods

There are 2 options to choose from.

  • lab: This method will determine the percentage vegetation present in each image by counting the ratio of detected green pixels.
  • deep: This method will make use of a deep image segmentation algorithm to derive percentage vegetation present in each image. (This is the preferred option).

Please see our paper for more details.

Example

# lab based segmentation
[phil@arasaka vegetation]$ algo run nocturne/vegetation/$(git rev-parse HEAD) -d \
	'{"src": "data://.my/test_images", "method": "lab"}'
Completed in 0.8 seconds
[0.0063,0.1172,0.0259]

# deep segmentation
[phil@arasaka vegetation]$ algo run nocturne/vegetation/$(git rev-parse HEAD) -d \
	'{"src": "data://.my/test_images", "method": "deep"}'
Completed in 8.7 seconds
[0.0087,0,0.0297]

Maintainer

About

(mirror) algorithmia vegetation service for UNGP

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published