AI for Palm Tree Health Diagnostics

Permia Sensing (UK-based SME/Imperial College London spin-out) uses a ground breaking acoustic sensor to monitor pest issues in coconut palms. With over 11 million farmers managing 12 million hectares of coconut plantations globally, coconut production is crucial to the economies of developing countries, particularly in top producers like Indonesia, the Philippines, India, Sri Lanka, and Brazil.

Typically, each hectare holds ~177 palms, contributing to annual yields of 15 billion nuts, with 12 billion processed into copra for coconut oil.

Although a single tree can produce around 75 coconuts annually, there's room to double this productivity. The coconut product market, driven by the rising demand for coconut water, is expected to surge from $1.5 billion in 2018 to $31.1 billion by 2026\. In Sri Lanka, smallholder farmers dominate the industry, affecting the socioeconomic landscape significantly.

A coconut palm typically begins yielding 6 to 10-years after planting, with potential yields ranging from 20 to 100 fruits per tree per year.

To meet growing demand, precision agriculture approaches are needed to increase plantation yields and bridge the ~50% productivity gap between potential and actual production.

In this Innovate UK funded project, Permia Sensing will use audio data captured using our innovative, highly sensitive mobile acoustic sensor to develop an AI-based algorithm to identify wider tree health issues such as diseases affecting transpiration (Lethal Yellowing/Bud Rot/Root (Wilt)/Ganoderma Wilt/Coconut Leaf Blight), nutrient deficiencies and other pests (Rhinoceros Beetle/Coconut Mite) to inform irrigation, fertilisation and pest management protocols in a precision farming approach.

Permia Sensing's platform-as-a-service model will boost productivity by using AI and innovative sensors to identify and address yield-affecting factors in real-time, enhancing agricultural efficiency and supporting community livelihoods in coconut farming regions.