InsectID
The system enables field scouting teams to photograph suspected pests, receive ranked identification candidates with confidence scores, and compare observed counts against action thresholds to guide treatment decisions. All observations and treatment records stay on the device offline first, then sync, with results logged for regulatory audit.
The problem
Crop-protection advisors and scouting teams spend hours each day identifying pests from field observations, yet species identification from photographs remains a skill taking years to develop. When identification is slow, scout bottlenecks delay treatment decisions. Simultaneously, scouts record observations in notebooks or spreadsheets that become disconnected from the treatment decisions made hours or days later, fracturing the audit trail. Treatments are applied based on habit rather than documented action thresholds, making it impossible to demonstrate compliance with integrated pest management obligations that Member States are now implementing under the sustainable-use framework.
The product
InsectID is a mobile platform that captures field conditions, pest photographs and insect counts during scouting, then ranks candidate pest identifications with confidence scores to help scouts confirm species without deep expertise. The system keeps observations, identification results and treatment decisions linked in an audit log, while a threshold calculator ensures every treatment references the economic-injury levels and action thresholds authored by the client's own experts. An IPM protocol library connects each identified pest to approved biological, cultural and chemical treatment options, grounding recommendations in the client's practices. Pheromone trap counts and IoT sensor readings feed population alerts to guide preventive decisions. All data stays offline first on the device, syncing when connectivity returns.
What it does
Scouting capture
The scouting app captures the field and crop stage, photographs of suspected pests, insect counts, GPS location, and maintains an offline queue so scouts work uninterrupted and sync when connectivity returns.
Pest identification (pilot species)
A vision model trained on a short-list of regional pests ranks candidate identifications with confidence scores, helping scouts confirm species quickly without expertise.
Threshold calculator
Economic-injury levels and action thresholds authored by the client's experts are versioned within the system so users always reference the current decision rule for each crop and pest.
IPM protocol library
Biological, cultural and chemical treatment options for each pest are curated, expert-approved and linked to the identification, so recommendations are grounded in the client's own protocols.
Decision & treatment records
Every treatment decision—product, date, reason and outcome—is logged to provide an audit trail for regulatory inspection and to track what works locally over seasons.
Trap-count ingestion
Pheromone trap counts and IoT sensor readings can be imported manually or automatically, with trend alerts flagging population pressure to guide preventive decisions.
Problems it solves
- Pest identification from field photographs is a skill that takes years to develop and remains slow in the field, creating bottlenecks in daily scouting workflows.
- Scouts record observations in notebooks or spreadsheets, but these notes are not linked to the treatment decisions made hours or days later, breaking the audit trail.
- Treatments are often applied based on habit or advice, without documented action thresholds to justify the intervention and demonstrate compliance with IPM obligations.
- Individual scouts work in isolation; no regional picture exists of where pests are emerging, making it hard to allocate resources or warn of invasive threats.
Built to order by Gyrus Tech Labs. Every project starts with a short discovery so the product fits how your operation really runs. EU-hosted, with the source code and documentation in your accounts.