Welcome to NexGenMap’s documentation!

This project aims to establish the next generation of land use and land cover mapping and monitoring tools by leveraging Planet’s improved satellite imagery and Google Earth Engine’s computing platforms. Focusing on sixteen Areas of Interest across seven distinct ecosystems of Brazil, Argentina & Paraguay (for a total of 300,800 square kilometers), this project will build on, 3.7m per pixel monitoring capabilities to pioneer new analytical approaches for automated, timely, more precise and up-to-date land use / land cover classification and monitoring systems.

Content

• 1. Project
  • 1.1. History
• 2. Mosaic
  • 2.1. Production of mosaics
• 3. Application with Machine Learning
  • 3.1. Random Forest
• 4. Application with Deep Learning
  • 4.1. Machine Configuration
  • 4.2. dl-semantic-segmentation
  • 4.3. Workflow Execution (Forest toy data)
  • 4.4. Usages
• 5. Important links