/invasive

Vue.js webserver with python/mongodb backend for biologists to classify images of invasive species for future use in training a CNN.

Primary LanguageJavaScript

Invasive Species Classification

A tool to help biologists working with Michigan Aerospace annotate drone imagery of natural areas, working towards a goal of using neural networks to classify and tag invasive species in environments at risk. The fun parts are written in D3.js.

To Begin:

  • Clone the repository!
  • Navigate to server/.
  • Open an instance of MongoDB (mongod).
  • Ensure any images you want in the database are in static/images/.
  • Run python populate_database.py ./images/.
  • Run python server.py. A flask server should start.
  • In another process, navigate anywhere in the repo and run npm run dev.
  • Navigate to localhost:8080 and begin classifying!