Core product · AI crop and pest intelligence
Pezego is Mutus Tech's AI crop and pest intelligence product. It supports mobile pest detection, contextual field diagnosis and crop-health monitoring — running as a focused mobile app, or connecting with Parallel Farm so pest observations stay linked to field records, tasks and reports.
Where Pezego fits
Assess
Mobile pest observations and crop-condition photos enrich the field baseline.
Plan
12-month NPKS plan and N-rate sensitivity (Parallel Farm).
Execute
Pezego captures in-field pest identifications and observations linked to Parallel Farm Farm Tasks.
Monitor
NDVI, climate history and Pezego pest-pressure signals tracked across the season.
Report
Pest findings consolidated with operating records for agronomist and partner review.
Pezego provides pest detection, in-field diagnosis and crop-health observations — not standalone IPM prescription. Recommendations are interpreted with soil tests, crop context, management records and validated through agronomy partners where required.
The mobile app
Pezego runs on iOS and Android and is designed for low-connectivity rural areas. Most identification responses arrive in under 30 seconds.
See it in action
A short walk-through of the main features, plus a 30-second clip of real-time pest detection in the field.
Videos are embedded via youtube-nocookie.com — no tracking cookies are set until a video is played.
How it works
01
A leaf is chewed, an aphid colony is forming, an unfamiliar larva is on a stem. Open Pezego in the field.
02
The app sends your image to our vision model, anchored to your country's pest taxonomy. You get the pest name and confidence score in seconds.
03
Confirm crop and growth stage, drop a GPS pin and Pezego auto-fetches the local weather. Extension officers can verify or correct the identification.
04
A risk level and a region-specific IPM recommendation, drawn from AHDB, RR98 and partner agricultural guidelines.
Coverage
40+
Pests in the Ghana knowledge base — Fall Armyworm, stem borers, cassava mealybug, pod borer and more.
19
UK pests covered out of the box — grain aphid, pollen beetle, CSFB, slugs, potato cyst nematode and more.
10
Crops covered in Ghana with detailed growth-stage taxonomy — Maize, Rice, Cassava, Yam, Cocoa, Cowpea, Groundnut, Tomato, Pepper, Plantain.
38
Major pests covered across the platform; the PEZEGO IEEE IoT Journal paper documents the 38-pest / 10-crop UK coverage and the 10,000+ annotated images underpinning the model.
The knowledge base draws on AHDB encyclopaedia entries, the RR98 crop-stage management review, region-specific agricultural guidance and our own field data. Coverage figures are documented in the peer-reviewed PEZEGO IEEE IoT Journal paper. New pests, crops and regions can be added through partner programmes.
Two regions, two playbooks
Focused on Fall Armyworm in Maize and the broader smallholder pest set, with a Ghana extension-officer network in place. Region (Greater Accra, Ashanti, etc.) and district selectors built in. Recommendations are written in clear, short sentences for low-input farms.
Tuned for UK agronomy. The recommendation engine references AHDB sources and the RR98 crop-stage management review, with structured data on insecticide thresholds and resistance.
Two audiences
Designed for use in the field on a phone, including by smallholders with limited data access.
Pezego is the front line of our pest-outbreak monitoring. ESOs see every record in their region, can verify and correct identifications and act as the trusted human-in-the-loop.
Beyond identification
Every observation carries a GPS pin. Officers and researchers can pull every report within a chosen radius, making early detection of regional outbreaks a click rather than a phone tree.
ESO verification flags create a layer of trusted data on top of crowdsourced reports — useful for monitoring, research and reporting to funders.
Field staff share findings, photos and treatment notes, tagged by pest and crop. A living, regional IPM playbook contributed by the people closest to the fields.
Peer-reviewed
IEEE IoT Journal · Volume 12, Issue 23
Co-developed with RSK ADAS, the University of Sheffield and the University of Liverpool. 20+ farms, 10,000+ annotated images, 38 pests, 10 crops.
Independent evaluation
+15%
Diagnostic accuracy
Improvement in pest diagnostic accuracy across a structured evaluation of 10 pest scenarios, with AI providing more context-specific advice than standard reference resources.
−20%
Processing time
Reduction in time-to-recommendation for the same pest scenarios, with refined interpretation of ambiguous pest descriptions.
Evaluation carried out under the BridgeAI / Innovate UK programme with an Independent Scientific Advisor paired through The Alan Turing Institute — Professor Po Yang of the University of Sheffield.
Source: Innovate UK Business Connect — Mutus Tech case study (April 2026) ↗
Get Pezego
Co-developed with the University of Sheffield, University of Liverpool and RSK ADAS. Available now on iOS and Android.
