Kelp Wanted: Segmenting Kelp Forests

Help researchers estimate the extent of Giant Kelp Forests by segmenting Landsat imagery. #climate

$15,000 in prizes
feb 2024
671 joined


Complimentary License and Benchmark

MathWorks, makers of MATLAB and Simulink, is sponsoring this challenge and the bonus award for Top MATLAB User. They're also supporting participants by providing complimentary software licenses, starter code, and learning resources.

To request your complimentary license, go to the MathWorks site, click the “Request Software” button, and fill out the software request form. This page also has several learning resources available for you.

MATLAB starter code to help you start working with this competition dataset quickly is available here!

During the course of the competition, the MathWorks Student Programs team is also available to support participants. Grace Woolson will be your point of contact for this competition. If you have questions, post it on the DrivenData community forum and someone from MathWorks or DrivenData will get back to you!

Additional resources

Below are some links to learn about MATLAB, Deep Learning Toolbox, and other useful resources for this competition.


  1. MATLAB Onramp
  2. Deep Learning Onramp
  3. Computer Vision Video Tutorials

General Resources

  1. Deep Network Designer App
  2. Hyperspectral Viewer App
  3. 9 MATLAB Cheat Sheets for Data Science
  4. Using Image Datastores
  5. Specify Layers of Convolutional Neural Network
  6. Define Custom Deep Learning Layers


  1. Train Simple Semantic Segmentation Network in Deep Network Designer
  2. Explore Hyperspectral Data in the Hyperspectral Viewer App
  3. Classify Hyperspectral Image Using Support Vector Machine Classifier
  4. Classify Hyperspectral Images Using Deep Learning


  1. Image Processing, Computer Vision and Deep Learning for Aerial and Satellite Images
  2. Deep Learning with MATLAB