McLaren Precision Agriculture
MPA Field · Basingstoke, Hampshire

Weed detection from your own imagery

You fly it. We turn it into a prescription.

Send us the multispectral captures from your Mavic 3M and a field boundary. You get back a weed map, a spray prescription and a survey report — checked by a person before it leaves here.

What makes this different

Every detection has a reason you can read

Most weed detection available today is built on trained AI models, which learn what a weed looks like from example images. That's a sound approach, and it works best when your field looks like the ones it learned from.

We've taken a different route. Detection here is deterministic: measured spectral thresholds across the green, red, red-edge and near-infrared bands, with spatial cleanup on top. Every polygon in your output can be traced back to the specific index values that produced it, and the same imagery run again next year gives the same answer.

That's useful beyond the agronomy. Spray records for herbicides and other plant protection products have to be kept for at least three years, and we keep your orthomosaics and maps for the same period. A map that can be explained in year three backs up that record — for an assurance audit, or when you're checking whether a patch has spread.

And a person looks at every job before it goes out. We see that as part of the service, not a stopgap.

The process

Three steps, and one of them is yours

STEP 01

Fly to the spec

Mavic 3M with RTK, at 30 m, 80/70 overlap, nadir. Cloud is fine. The full spec is below — it's short, and it's a gate, not a suggestion.

Your time in the field
STEP 02

Upload and get checked

Send the captures, your field boundary, and the product or tank mix and rate you plan to apply. We check the upload before you're charged anything. If it won't process, we tell you why and what to change — and there's no charge.

Same day
STEP 03

Maps come back

Orthomosaic, index maps, classified weed raster, patch statistics and a prescription you can load straight into the sprayer terminal.

Within 48 hours

Flight specification

What the imagery has to be

There's no weather window here, no calibration panel and no time-of-day rule. Detection looks at the relative difference between weed and crop within your flight, the Mavic 3M's sunlight sensor corrects for changing light frame by frame, and thresholds are set for each field. What the spec protects is the three things processing can't fix afterwards: geometry, a complete band set, and RTK position.

Required — checked on every upload
ParameterRequirement
SensorDJI Mavic 3M
BandsG · R · RE · NIR
PositioningRTK fixed ≥ 95%
Sunlight sensorFitted, unobstructed
Altitude25–40 m AGL
Forward overlap≥ 80%
Side overlap≥ 70%
Camera angleNadir (−90°)
Elevation optimisationOptional
Field boundary.shp + .dbf + .shx

Sent back, not charged

  • Positioning not RTK fixed — include the .MRK file from each flight folder
  • RGB-only imagery — the index maps need all four bands
  • Two sessions merged into one upload
  • Fewer than 20 frames, or more than 6,000
  • No field boundary

We'd rather reject an upload than process it badly and explain afterwards. A rejection comes with the reason and the fix.

G
560 ± 16 nm
Primary matching band. GNDVI numerator.
R
650 ± 16 nm
Chlorophyll absorption. NDVI, LCI.
RE
730 ± 16 nm
Red edge. NDRE, canopy condition.
NIR
860 ± 26 nm
Structure and biomass. All four indices.
Ground resolution by flight altitude
AGLMS GSDAccepted
25 m1.15 cmYes — finest
30 m1.38 cmYes — recommended
40 m1.84 cmYes
50 m2.30 cmBy arrangement
60 m2.76 cmNo

Everything is processed and delivered at native resolution. That's the point: at 30 m the Mavic 3M resolves individual rye grass tillers inside a wheat canopy, and averaging down would throw away exactly the small, early patches you most want to catch.

RTK is required because a prescription has to be right in absolute position, not just relative. Your sprayer places each nozzle by RTK; a map on standard GNSS is out by metres, which puts a 25 cm spray buffer on clean crop and misses the weed.

Deliverables

What lands in your inbox

GeoTIFF
Calibrated orthomosaic and index maps — NDVI, GNDVI, NDRE and LCI at native resolution, georeferenced and ready for QGIS.
GeoTIFF
Classified weed raster — detected weed, non-candidate vegetation, and out-of-boundary, as a single-band mask.
Shapefile
Patch polygons — one feature per detected patch, with area, size class and the mean index values behind the call.
Shapefile
Spray prescription — spray and no-spray zones with a GPS buffer sized to your sprayer's nozzle or section resolution, carrying your product and rate so it loads ready to go.
ISOXML
Application map — for direct upload to an ISOBUS terminal. John Deere Operations Center upload confirmed working.
CSV
Patch statistics — per-patch area, centroid and index means, so the numbers in the report can be checked independently.
PDF
Survey report — detection statistics, patch distribution, coverage notes, the thresholds used, and modelled treated-area reduction by sprayer type.

Evidence

What we've measured, and what we haven't

83%
Treated-area reduction, measured from the sprayer's as-applied record in John Deere Operations Center.
Hampshire arable farm · Sep 2026
1 day
Survey to spot spray. Volunteer cereals and mixed weeds, treated with glyphosate.
Same field
10 days
Follow-up visit: the weeds in the sprayed zones were dead.
Same field · photographed

The field was flown, the prescription went into the sprayer the next day, and the as-applied data came back from John Deere Operations Center: the sprayer treated 17% of the field.

Measured, not modelled. That 83% comes from what the sprayer actually did, not from the geometry of our map. Any figure we haven't got as-applied data for is labelled as modelled — and it's a fair question to ask of any savings figure you're quoted.

Prescription maps are accurate at the point of survey. Weeds that emerge after the flight aren't on the map. On that field a fresh flush came up between the survey and the glyphosate taking effect — that's UK weather, not a detection miss, but it's the reason to fly as close to your spray date as you can.

One more thing worth saying plainly: thresholds are set per field and per growth stage. A new field means a calibration step. It takes a fraction of the time the original development did, and it's what the person in the loop does on every job.

Pricing

Per hectare, on the boundary you supply

Weed detection and prescription

£12 / ha
  • Orthomosaic and index maps at native resolution
  • Weed map, patch statistics and survey report
  • Prescription with your product and rate, ready for the terminal
  • A call to walk through the map before you commit to a pass

Drone contractors

Let's talk

Processing fields for your own customers? Volume rates and unbranded delivery are agreed directly.

Ask about contractor rates →

You're invoiced once your upload passes our checks, and can pay by card or bank transfer. For a first job the maps are sent as soon as payment comes through — card payment means that can be minutes, not days. Regular customers can move to monthly invoicing.

Rejected uploads are free. If a job passes our intake check and then fails to process, that's our problem to fix or refund, not yours. What we don't refund is agronomic disagreement — the map tells you what was spectrally detectable from the air on the day, which is an input to your decision rather than the decision.

Questions

The ones people actually ask

Why 48 hours?
Building a calibrated orthomosaic from a few hundred multispectral frames takes hours of compute, and every map is checked before it goes out. Turnaround figures quoted in minutes usually refer to redrawing a prescription from an orthomosaic that's already been built. We quote 48 hours and usually come back inside 24.
Do you tell me what to spray?
No. We map where the weeds are; what goes in the tank, and at what rate, is your call or your agronomist's. Tell us the product or tank mix and the rate and we'll write them into the prescription, along with the total product the map implies you'll need.
Do I need to wait for sunshine?
No. This is the UK — if we only processed sunny flights we'd process nothing. Detection works on the relative difference between weed and crop within your flight, and the Mavic 3M's sunlight sensor corrects for light changing as cloud comes and goes. The one thing we do look for is a cloud shadow sitting on part of the field while the drone is in sun, which the sensor can't see; if it lands on the detection area we'll tell you.
Do you keep or train on my imagery?
Raw imagery is kept for 90 days so we can re-run a job if you query it, then deleted. The orthomosaic, index maps and prescription are kept for three years — the same period as your spray records — so you can ask for a past map or compare seasons. We don't train models on your data, and we never share or sell your imagery, maps or field data. The only thing that leaves here is anonymised, aggregated statistics — weed pressure by crop and region, for example — that can't be traced back to your farm or field. If you want anything deleted sooner, say so and it goes.
What if my drone isn't a Mavic 3M?
Ask. Altum, RedEdge and P4 Multispectral are all workable in principle — it's a question of band mapping and calibration chain, and we'd want to run one field through before quoting you a rate. What won't work is RGB-only imagery from any aircraft: without the near-infrared band there are no index maps to threshold.
Can you map species X?
Proven so far on volunteer cereals and mixed weeds for a glyphosate spot spray, and on thistle and rye grass in growing cereals. Blackgrass and docks are the obvious next ones and we're actively looking for fields to characterise. If your target isn't on the list we'll say so rather than take the money and hope.
Will this work with my sprayer?
The prescription is delivered with a buffer sized to what your sprayer can actually resolve — individual nozzle control at 25 or 50 cm, or boom section control at whatever your section width is. Tell us which you have. Section control at 4 m gets a meaningfully smaller reduction than nozzle control, and the report models all three so you can see the difference.
Who's actually doing the work?
Craig McLaren. MPA is a small operation and the processing is done here in Hampshire, on our own hardware, by the person who wrote the detection software. You'll be talking to him, not a support queue.

No drone?

You don't need one to get a weed map

For farms

We'll fly it for you

MPA can fly the field with our own Mavic 3M and run it through the same process. Survey and processing are quoted together, depending on where you are and how much ground there is.

Ask about a survey →
For drone pilots

Join the pilot network

We're building a network of pilots who can fly fields to this spec, so farms further afield can get mapped without the travel. If that sounds like you, we'd like to hear from you. You'll need:

  • A Mavic 3M with RTK
  • The CAA permissions and insurance for commercial work
  • To be happy flying to the spec above
Register your interest →

Get started

Send one field and see what comes back

The sensible first job is a single field you already know well, so you can check the map against what's actually out there. Email for an upload link and the job sheet.