agriQ

Our Mission

Sustainable Crop Management

Food security and environmental protection must go hand-in-hand. By translating aerial data into field-level insights, agriQ helps farmers transition to a truly precise, data-driven operational model.

Optimise Resource Allocation

Pinpoint exactly where water and fertilizer are needed across your entire farmland, eliminating waste and lowering overhead costs.

Reduce Chemical Inputs

Apply targeted treatments only where disease risk is explicitly identified, protecting the surrounding ecosystem and soil health.

Secure Yields

Use advanced data-driven forecasts to make proactive decisions that secure and improve your harvest, season after season.

Prevent Crop Disease

Detect early signs of blight, fungal disease, and nutrient stress weeks before they are visible, enabling targeted treatment before losses occur.

How It Works

Our Project

Modern farming demands more than intuition. agriQ's four-step pipeline turns raw spectral data into crop health maps, disease alerts, and yield forecasts so every decision is backed by evidence, not guesswork.

Satellite Positioning

Multispectral Imaging

Step 01

Field Scanning

Drone flights are conducted over farmland to capture detailed aerial imagery and multispectral data across the entire field, providing a comprehensive digital picture of crop conditions at scale.

Multispectral Imaging Full Coverage High Resolution Real-time Data

NDVI 0.74

NDRE 0.41

CWSI 0.18

LAI 3.2 m²/m²

Chl 38 μg/cm²

● SCAN ACTIVE Field: 23.4 ha · 4s/rev

Step 02

Crop Health Analysis

Using spectral data and advanced machine learning, we identify variations in plant health, nutrient stress, and potential disease risks before they are visible to the naked eye.

NDVI Analysis Disease Detection Nutrient Mapping

crop_analysis.py

import numpy as np

import torch, rasterio

from pathlib import Path

from dataclasses import dataclass, field

from typing import Dict, List, Tuple

# ─── Spectral indices ───────────────────

def ndvi(nir: np.ndarray, red: np.ndarray) -> np.ndarray:

return (nir-red) / (nir+red+1e-6)

def ndre(nir: np.ndarray, re: np.ndarray) -> np.ndarray:

return (nir-re) / (nir+re+1e-6)

def cwsi(v: float, t=0.3) -> float:

return float(np.clip(1-v/t, 0, 1))

# ─── Data structures ────────────────────

@dataclass

class FieldZone:

zone_id: str

ndvi: float

ndre: float

cwsi: float

risk: str = "nominal"

alerts: List[str] = field(default_factory=list)

# ─── Analyser class ─────────────────────

class CropHealthAnalyser:

def __init__(self, model_path: Path):

self.model = torch.load(model_path)

self.model.eval()

self.zones: List[FieldZone] = []

def load_bands(self, p: Path) -> Dict:

with rasterio.open(p) as src:

b = src.read().astype(np.float32)

return {"r":b[0],"g":b[1],"nir":b[3],"re":b[4]}

def analyse(self, p: Path) -> List[FieldZone]:

bd = self.load_bands(p)

nv = ndvi(bd["nir"], bd["r"])

nr = ndre(bd["nir"], bd["re"])

cw = cwsi(float(nv.mean()))

masks = self._segment(nv)

for i, m in enumerate(masks):

zn = float(nv[m].mean())

zr = float(nr[m].mean())

rk = "high" if zn < 0.25 else "low"

z = FieldZone(f"Z{i:02d}", zn, zr, cw, rk)

if rk == "high":

z.alerts.append("Disease risk")

self.zones.append(z)

return self.zones

def _segment(self, ndvi_map):

# Spatial clustering on NDVI surface

t = torch.from_numpy(ndvi_map)

return self.model(t.unsqueeze(0))

def prescription(self) -> Dict:

return {z.zone_id: z.risk for z in self.zones}

# ─── Run ────────────────────────────────

if __name__ == "__main__":

a = CropHealthAnalyser(Path("model.pt"))

res = a.analyse(Path("field.tif"))

pmap = a.prescription()

for zid, rk in pmap.items():

print(f"{zid}: {rk}")

import numpy as np

import torch, rasterio

from pathlib import Path

from dataclasses import dataclass, field

from typing import Dict, List, Tuple

# ─── Spectral indices ───────────────────

def ndvi(nir: np.ndarray, red: np.ndarray) -> np.ndarray:

return (nir-red) / (nir+red+1e-6)

def ndre(nir: np.ndarray, re: np.ndarray) -> np.ndarray:

return (nir-re) / (nir+re+1e-6)

def cwsi(v: float, t=0.3) -> float:

return float(np.clip(1-v/t, 0, 1))

# ─── Data structures ────────────────────

@dataclass

class FieldZone:

zone_id: str

ndvi: float

ndre: float

cwsi: float

risk: str = "nominal"

alerts: List[str] = field(default_factory=list)

# ─── Analyser class ─────────────────────

class CropHealthAnalyser:

def __init__(self, model_path: Path):

self.model = torch.load(model_path)

self.model.eval()

self.zones: List[FieldZone] = []

def load_bands(self, p: Path) -> Dict:

with rasterio.open(p) as src:

b = src.read().astype(np.float32)

return {"r":b[0],"g":b[1],"nir":b[3],"re":b[4]}

def analyse(self, p: Path) -> List[FieldZone]:

bd = self.load_bands(p)

nv = ndvi(bd["nir"], bd["r"])

nr = ndre(bd["nir"], bd["re"])

cw = cwsi(float(nv.mean()))

masks = self._segment(nv)

for i, m in enumerate(masks):

zn = float(nv[m].mean())

zr = float(nr[m].mean())

rk = "high" if zn < 0.25 else "low"

z = FieldZone(f"Z{i:02d}", zn, zr, cw, rk)

if rk == "high":

z.alerts.append("Disease risk")

self.zones.append(z)

return self.zones

def _segment(self, ndvi_map):

# Spatial clustering on NDVI surface

t = torch.from_numpy(ndvi_map)

return self.model(t.unsqueeze(0))

def prescription(self) -> Dict:

return {z.zone_id: z.risk for z in self.zones}

# ─── Run ────────────────────────────────

if __name__ == "__main__":

a = CropHealthAnalyser(Path("model.pt"))

res = a.analyse(Path("field.tif"))

pmap = a.prescription()

for zid, rk in pmap.items():

print(f"{zid}: {rk}")

Step 03

Actionable Insights

The collected data is translated into clear field maps and zone indicators that highlight areas requiring attention, exportable at any time through the platform.

Field Zone Maps Risk Alerts Export Ready

Field Map · Zone Analysis NDVI scan active

NDVI

0.61

STRESS

18%

ALERT

3 zones

Healthy

Stress

Alert

Low vigour

Step 04

Smarter Farm Decisions

Farmers and agronomists can use these insights to optimise irrigation, fertilisation, and crop protection throughout the season, with all data readily accessible and exportable via the app.

Irrigation Planning Fertilisation Maps Season Reports

Crop Health Map

Real-time

Stress Detection Alerts

Active

Field Performance Insights

Downloadable

Early Disease Prevention

Predictive

Yield Projection

Seasonal

Prescription Maps

Zone-based

AI Models

Our Crop Models

We develop specialised models tailored to analyse different crops, starting with wheat and potatoes.

Wheat

Qwheat

Our multispectral drone intelligence platform analyses a wheat field from the air and produces a complete agronomic prescription, identifying where nitrogen is deficient, where water stress is developing, where disease is emerging, and how many heads the crop is carrying.

Wheat head detection across the field

Nitrogen deficiency alerts

Biomass and growth monitoring

Yield estimation for harvest planning

Potatoes

Qpotato

Potato crops require close monitoring throughout the growing season. Our model is trained specifically on potato disease prediction, using multispectral drone imagery to detect early signs of late blight, early blight, and other fungal diseases before visible symptoms appear. It also identifies water stress and nutrient imbalances across the field.

Disease prediction (late blight, early blight)

Water stress detection

Nutrient deficiency identification

Field variability mapping

App Demo

Qsense. Your field.
Your data. Your decisions.

See how our platform turns a single drone flight into field-level intelligence, delivered straight to your device.

Why It Matters

A growing crisis
demands precision

Research and field data make clear why drone-based crop intelligence is no longer optional. It is essential.

Food Security

World's Wheat Supply at Risk from Heat and Drought

A study warns that simultaneous heat and drought events driven by climate change could deliver dangerous shocks to global wheat supply, with concurrent crop failure risks in major producing nations rising sharply.

NBC News · 2023 Read more

Production

FAO Lowers 2024 Global Wheat Forecast to 791 Million Tonnes

The FAO cut its 2024 global wheat production forecast due to reduced EU plantings and dry conditions, with North Africa continuing to face severe drought, reinforcing the fragility of global supply chains.

Miller Magazine · 2024 Read more

Climate

Climate Change Cuts Global Crop Yields, Even When Farmers Adapt

A Nature study found global staple crop yields could decline by around 24% by 2100 under high-emissions scenarios, with global food production falling roughly 4.4% per 1 degree C of warming.

Stanford / Nature · 2025 Read more

Precision Ag

Precision Agriculture Helps Farmers Navigate Climate Challenges

Precision agriculture tools including variable-rate application and crop monitoring enable farmers to cut waste and optimise inputs, improving profitability while reducing environmental impact.

Farming First · 2024 Read more

Drones

From Sky to Soil: How Drones Are Revolutionising Crop Monitoring

A two-year CGIAR study found that drone-based remote sensing with multispectral sensors and machine learning can detect plant stress before symptoms are visible, enabling more targeted management.

CGIAR · 2025 Read more

Efficiency

Why Precision Agriculture Is Essential in Combating Climate Change

The WEF estimates that 15 to 25% farm adoption of precision agriculture could boost global yields 10 to 15% while reducing greenhouse gas emissions by 10% and water use by 20% by 2030.

AgFunderNews · 2025 Read more

Why agriQ

The key benefits of our
technologies for modern farming

Drone-based crop intelligence helps farmers monitor health, detect risk early, and optimise resources for more productive and sustainable agriculture.

Crop Health Monitoring

Multispectral imagery reveals crop stress, nutrient deficiencies, and plant health variations across fields before they are visible to the eye.

Early Disease Detection

Identify early signs of disease and pest pressure using spectral data analysis, allowing intervention before problems spread.

24/7 Availability

Access crop health data anytime through our platform, allowing monitoring of field conditions throughout the entire growing season.

Cost Reduction

agriQ helps farmers reduce unnecessary fertiliser, pesticide, and labour costs while improving overall field efficiency.

Large-Scale Field Scanning

Drone flights efficiently cover large areas of farmland, providing comprehensive field insights impossible to gather manually.

Yield Monitoring and Forecasting

Analyse crop growth patterns throughout the season to estimate yields and support better harvest planning.

Who We Are

A team built
at Imperial

agriQ is a team of scientists and engineers from Imperial College London driven by a shared vision for sustainable agriculture. We develop innovative technologies to address environmental and food security challenges, helping farmers optimise resources, improve crop health monitoring, and make more informed decisions about their land.

We are committed to building solutions that empower the people who feed the world.

Innovation for Agriculture

We are committed to developing new technologies that help farmers better understand and manage their fields.

Committed to Integrity and Trust

Trust and transparency are at the core of everything we do for our clients and partners.

Sustainability

We believe modern farming should increase productivity while protecting soil, resources, and the environment.

Putting Farmers First

Your success is our priority. We build solutions that make a real, measurable impact on the ground.

Our Team

Mechanical Engineering

Early Access

Join our waitlist

Be among the first farms to experience what precision aerial intelligence can do for your yields and your land. Whether you are a farmer, agronomist, or working in the field, we would love to hear from you.

Or contact our team directly to learn how agriQ can bring precise aerial intelligence to your fields.

Aerial intelligencefor sustainableagriculture