Anyone here done crop yield prediction using RGB drone images + ML?
Hey guys,
I'm currently working on my capstone project and was wondering if anyone here has experience with using drones for crop analysis/yield prediction.
Our plan is to use an **RGB drone** (can't afford a multispectral/NIR one 😅), stitch the images into an orthomosaic, extract features like **VARI**, canopy coverage, plant density, etc., then feed those into a machine learning model (probably Random Forest) to predict **rice yield**.
Has anyone here tried something similar?
* Does using only RGB imagery actually work well enough for yield prediction? * Is VARI useful, or are there better RGB indices/features I should look into? * Any tips, things you wish you knew before starting, or common mistakes to avoid? * If you've done this before, how accurate were your results?
We're just trying to estimate yield from drone imagery.
Would really appreciate any advice, papers or just hearing about your experience. Thanks! #technology source