The context
Measuring whether a spray treatment properly covered foliage has traditionally been a lab exercise or relied on physical water-sensitive cards: slow, manual, and hard to apply at field scale. The agronomist ends up deciding on a hunch rather than a number — meaning wasted chemical when over-applied, or insufficient coverage when under-applied.
The technical solution
Spray Analyzer solves this directly: the treated leaf is illuminated with a standard 365nm UV lamp — the spray's dye fluoresces under that light — and photographed with a normal phone camera. The analyzer's computer vision model isolates the leaf in the image, classifies each pixel by whether it fluoresces, and calculates the exact coverage percentage, leaf by leaf.
The entire process — from photo to number — happens in seconds, with no lab, no spray cards and no additional software installation required.
The result
Figures published by the product itself at sprayanalyzer.com, not our own estimates.
In real production
Spray Analyzer isn't a lab prototype: it's deployed and in daily use with Almagrícola, a real farming operation in Bogotá, Colombia.
"Coverage went from a gut feeling to a number on screen. We apply smarter and cut chemical use smarter."
Testimonial originally published at sprayanalyzer.com.
Technology used
Computer vision with image segmentation and fluorescent-pixel classification, trained specifically for this case — not a generic model. The same engineering approach (Python, vision models, cloud deployment) we apply to any computer vision project at FutureSynap.
Why it's our central case
Spray Analyzer is PlusSeed's project that best demonstrates what FutureSynap brings to the European market: computer vision applied to a real business problem, with verifiable data and a real client using it in production — not a promise, not a demo. It's the yardstick we measure every new project against.