fairfolio
Saved Projects

Mapping Invasive Pampas Grass Through Drone-Based Photography and Machine Learning

CSEF · 2026 Earth & Environmental Sciences(Senior Division)

Andy Feng

Overview

This project aims to use drone-based aerial imaging to implement an Artificial Intelligence model to detect and map pampas grass on a large scale. To experiment, I went to the field and collected data on Pampas Grass across San Diego, including at Miramar Lake and Rose Canyon. I then programmed and trained an AI algorithm based on the YOLO AI model in Python. Using hundreds of images, the algorithm was trained to locate Pampas Grass clumps from drone-taken images correctly. From all the photos, 70% were used for training the model, 20% to validate, and 10% to test. Testing was split into two iterations to show improvement in results. The first iteration showed some success but also revealed errors. The second iteration used more photos in the training set and had a higher rate of success of 88%, correctly mapping the majority of Pampas Grass. The results indicate that combining the speed and large-scale capture from a drone with the accuracy of the AI model proves the process is highly effective in detecting and mapping Pampas Grass by using repeated learning from observing physical features, such as color and shape. However, possible errors may have occurred due to the low error rate of the AI model. In summary, the methodology used is highly effective in detecting invasive plants and has massive potential for aiding weed removers and conservationists. By successfully finding Pampas Grass automatically, the step of manually finding Pampas Grass can be skipped, leaving more time for removing the weeds and restoring native ecosystems around San Diego.

Competition history

  • CSEF 2026 Earth & Environmental Sciences(Senior Division) · Entry S-08-21

Related projects

Closest projects by meaning, across every fair and year in the corpus.

Browse more like this

Source: California Science & Engineering Fair public projects

Save projects to your library

Sign in with Google to keep track of projects you find interesting, organized into folders. An account also raises your daily allowance for “Has this been done?”, and lets you create a key for the MCP server with a much higher limit than anonymous use. Browsing stays public.

Continue with Google