Notebooks

We present a complete list of the notebooks available in FarmVibes.AI with a short summary for each of them. Besides their description, we also include the expected disk space and running time required per notebook, considering the recommended VM size.

Summary

We organize available notebooks in the following topics:

Getting Started

ADMAg - Microsoft Azure Data Manager for Agriculture

BingMaps

Segment Anything Model Workflow (BingMaps basemap) 📓

Crop Analysis

Crop Land Segmentation

Deforestation

Index Computation

Landsat

Model Training

ONNX Integration

Remote Sensing

Segment Anything Model

Sentinel

SpaceEye

Sustainability

Time Series

Harvest and germination periods 📓

Water Management

Weather

What-if Analysis

Working with Custom Workflows

Notebooks description

Carbon sequestration evaluation with Microsoft Azure Data Manager for Agriculture (ADMAg) and COMET-Farm API 📓 : Derive carbon sequestration information with COMET-Farm API for agricultural fields, using Azure Data Manager for Agriculture to retrieve farming data.
Crop cycle detection 📓 : Run time-series analysis over NDVI data to detect the number of crop cycles in a year for a certain region.
Crop land segmentation (1/4) - dataset generation 📓 : Generate a dataset based on NDVI and CDL rasters to train a crop land segmentation model.
Crop land segmentation (2/4) - dataset visualization 📓 : Visual exploration of the crop land segmentation dataset.
Crop land segmentation (3/4) - local model training 📓 : Train a crop land segmentation model locally.
Crop land segmentation (3/4) - model training on Azure Machine Learning 📓 : Train a crop land segmentation model on Azure Machine Learning.
Crop land segmentation (4/4) - inference 📓 : Infer crop land segmentation for new regions with a trained model.
Detecting Forest Changes 📓 : Helps users to detect forest changes
Download ALOS forest extent maps 📓 : This notebook downloads the ALOS (Advanced Land Observing Satellite) forest extent maps
Download Glad Forest Map 📓 : This notebook downloads the Global Land Analysis (GLAD) forest extent maps.
Download Global Forest Change (Hansen) maps. 📓 : This notebook contains functions to download and process the Global Forest Change (Hansen) maps.
Field boundary segmentation (SAM exploration) 📓 : Segment Anything Model exploration over FarmVibes.AI data to segment crop field boundaries.
Field-level Irrigation Classification 📓 : Estimate an irrigation probability map over crop fields segmented with Segment Anything Model.
Field-level spectral indices 📓 : Compute and visualize spectral indices over crop fields segmented from Sentinel-2 and SpaceEye imagery.
Green House Gas fluxes 📓 (30s) : Compute GHG emissions and sequestration for a given crop, location and time range.
Harvest and germination periods 📓 : Infer harvest and germination periods from NDVI time series.
Hello World 📓 : Simple example of how to instantiate a client and run a workflow.
Integration with ADMAg to compute NDVI summary 📓 : Connecting FarmVibes.AI and Microsoft Azure Data Manager for Agriculture (ADMAg)
Investigating Amazon Rainforest deforestation with SpaceEye 📓 : Download and preprocess Sentinel data to obtain cloud-free images using SpaceEye over the Amazon Rainforest.
Irrigation classification 📓 : Generate irrigation probability map by running workflow that uses Landsat8 image and DEM data
Land degradation 📓 : Compute NDVI trends on Landsat images and analyse precipitation data to detect land degradation.
Micro climate prediction 📓 : Train a model to predict micro climate parameters given global weather information and local weather station data.
Nutrient Heatmap Estimation - Classification 📓 : Using soil samples from a farm, train a Random Forest model to estimate a heatmap of soil nutrients.
Nutrient Heatmap Estimation - Neighbors 📓 : Estimate a heatmap of soil nutrients based on neighboring soil samples with location suggested by an optimal sensor placement workflow.
Nutrient Heatmap Estimation with ADMAg integration 📓 (~1GB, <1 minute) : Using soil samples from a farm retrieved with ADMAg, train a Random Forest model to estimate a heatmap of soil nutrients.
Optimal Sensor Placement 📓 : Helps to find optimal sensor placement location using satellite imagery. The optimal locations identified can be used to collect soil properties by installing sensors or performing other manual operations
Optimal Sensor Placement over Segmented Crop Field 📓 : Segment a field with Segment Anything Model over Sentinel-2 imagery and estimate optimal sensor placement for the segmented region.
Segment Anything Model Workflow (BingMaps basemap) 📓 : Segment BingMaps basemaps with SAM workflow.
Segment Anything Model Workflow (Sentinel-2) 📓 : Segment Sentinel-2 imagery with SAM workflow.
Spectral indices 📓 : Compute and visualize spectral indices from Sentinel-2 imagery.
Timelapse visualization 📓 : Generate a timelapse of SpaceEye and NDVI rasters.
Weed detection 📓 : Train a Gaussian Mixture Model to group similar regions in a raster image affected by weed.
What-if analysis of carbon sequestration 📓 : Derive carbon sequestration information with COMET-Farm API for agricultural fields.