Spot-spraying vision system
Inspired by publicly reported work at John Deere
Detect weeds in field images in real time and decide which nozzles to open, then estimate savings and risks.
The brief
John Deere has described See & Spray: boom-mounted cameras, a model that detects weeds, and nozzles that open only where a weed is seen, with reported herbicide savings. Build a spot-spraying system of your own: segment crop and weed in field images, map detections to nozzle zones as the boom moves, run it fast enough for a moving sprayer, and estimate savings and missed weeds. Good looks like: segmentation IoU for crop and weed, frames per second on modest hardware, a nozzle simulation with savings and misses, and a field trial plan.
Read the case first
What John Deere has said in public about this kind of work, on its own pages:
Make it your own
Design it for small Indian farms: a backpack sprayer or a drone instead of a large boom, and a phone instead of a computer on the tractor. Show what changes in speed, cost and accuracy.
AITeky has no link with John Deere, which has not reviewed or endorsed this brief. The brief only draws on what the company has published, uses open data rather than its data, and is a learning exercise: don't present your project as John Deere's work or system.
What you will build
- Train a crop/weed segmentation model
- Measure IoU and speed on modest hardware
- Map detections to nozzle zones for a moving boom
- Simulate a field pass: herbicide used and weeds missed
- Compare with spraying the whole field
- Write a field trial plan with safety checks
Free, open data to build it with
Sign up free to get the data and links
It takes a minute with your email. AiTeky Pass (₹999/month or ₹9,999/year) is free for your first 30 days.