Webinar: From Citizen-Science to Your Phone: Using iNaturalist's 70M+ Crowdsourced Images to Build an ML Insect Identification App

Image
Individual post

When

11 a.m., Nov. 18, 2022

Webinar Materials

 

Presentation Slides

 

About the Webinar

iNaturalist is a popular, citizen-science platform that contains an image database of 70+ million research-grade, labeled, open source images of living organisms: an incredibly rich resource for natural historians and scientists, from amateurs to professionals. To enable researchers to access this massive image database, collaborators Zi Deng of the University of Arizona and Shivani Chiranjeevi of Iowa State University will demonstrate iNaturalist Open Download, a python tool they've built to allow researchers, especially those training Machine Learning models, to easily query and download subsets of images from the iNaturalist Open Dataset.

Zi and Shivani will demonstrate a typical use case for their tool, and will then present an insect classification model that uses Self-Supervised (SSL) techniques that they are working on as collaborators in the AI Institute for Resilient Agriculture (AIIRA, based at Iowa State University) that utilizes data from the tool.

Image
naturalist logo

Aimed at researchers, this webinar will showcase an app that is primarily built for people interested in acquiring more data for their machine learning models. However, anyone interested in image datasets of living organisms will benefit from learning about iNaturalist Open Download. And if you're a farmer, plant scientist, entomologist, or just buggy for insects, join us for a demonstration of the mobile phone app Zi and Shivani have created that helps identify various insect-pest species in the field.

What You'll Learn

  • How to find and use the iNaturalist Open Download app to access millions of research-grade images available in iNaturalist Open Dataset
  • How to acquire more image data for training machine learning models
  • About an insect classification model that uses Self-Supervised (SSL) techniques on data accessed from the iNaturalist Open Download app 

About the Presenters

Zi Deng is a doctoral student in the University of Arizona's Electrical and Computer Engineering department and a Researcher at the UArizona's Data Science Institute. He received both his BS and MS in Electrical Engineering at UC San Diego, with a focus on signal and image processing. Zi is part of the collaborative, multi-institutional team that comprises the AI Institute for Resilient Agriculture, whose mission is to create new AI-driven, predictive digital twins for modeling plants and deploy them to increase the resilience of the Nation's agricultural systems.

Image
Shivani Chiranjeevi.

Shivani Chiranjeevi is a doctoral student in Mechanical Engineering at Iowa State University in Ames, Iowa. She specializes in Artificial Intelligence (AI), image processing, and computer vision in plant sciences. She received her BS in Computer Science from Iowa State, and, like Zi, is part of the AIIRA collaboration, but based at Iowa State.

Read more about AIIRA and its vision, activities, and collaborative projects.

 

 

Image
aiira logo

Create Account

An Open Science Workspace for Collaborative Data-driven Discovery