Spotting Crop Disease From a Single Phone Photo

Executive Summary

Agriculture supports the livelihoods of hundreds of millions of people across Asia, yet crop diseases continue to erode farm incomes every season. The earlier a disease is detected, the cheaper and more effective the treatment - but early detection depends on expertise that most farmers cannot access in time.

TokenWave AI developed an AI-powered plant disease detection system that turns an ordinary smartphone into a diagnostic tool. A farmer photographs an affected leaf, and within seconds the system identifies whether the plant is diseased, names the specific disease, shows how confident it is and recommends what to do next.

Plant diseases caused by fungi, bacteria and viruses spread quickly in favourable weather. A problem that starts on a few leaves can affect an entire field within days. Timely and correct identification is therefore the single most important factor in protecting a harvest.

In practice, diagnosis relies on the farmer's experience, advice from local input dealers or occasional visits from agricultural officers. With one extension worker often responsible for thousands of farmers, expert advice rarely arrives at the moment it is needed most.

Why Existing Approaches Fall Short

• Manual inspection depends heavily on individual experience and is inconsistent.
• Laboratory testing is accurate but slow, costly and far from most farms.
• Generic advice from dealers is often product-driven rather than diagnosis-driven.
• Simple image models trained on clean lab photos struggle with real field conditions.

What Was at Stake

• Yield loss: diseases left untreated for even a few days can significantly reduce harvest quality and quantity.
• Input costs: unnecessary or incorrect spraying increases cultivation cost and chemical residue.
• Farmer income: for small and marginal farmers, a single bad season can mean debt.
• Food security and sustainability: overuse of chemicals harms soil health and the environment.

We followed a structured, problem-first process - understanding the real-world problem before choosing the technology, and validating every stage before moving to the next.

An AI system that identifies plant leaf diseases from a single smartphone photo and recommends the right action - before the disease spreads across the field.

Solution Workflow

Key Capabilities

Technology & Techniques

“The goal was simple: put a reliable crop doctor in every farmer's pocket, so that the first sign of disease leads to the right action - not a lost harvest.”
— TokenWave AI Project Team

Who Can Benefit

• Farmers and FPOs for early, affordable diagnosis at the farm.
• Agri-input dealers for recommending the correct product based on an actual diagnosis.
• Agritech startups for adding intelligent crop health features to their apps.
• Agricultural departments and KVKs for monitoring disease outbreaks across regions.