Nur Sabrina Ahmad Azmi, Zarina Zainuddin, Nurul Hidayah Samsulrizal, Noor Haslinda Noor Hashim, Paramita Cahyaningrum Kuswandi, Anna Rakhmawati, Nur Aeni Ariyanti
Durian, the beloved "king of fruits", serves as an economic lifeline for Southeast Asia, yet it faces constant threats from fungal pathogens that can destroy up to 40% of a harvest. While traditional visual checks are often subjective and laboratory tests like polymerase chain reaction (PCR) are too complex for daily field use, machine learning (ML) offers a more supportive, practical path forward. By using convolutional neural networks (CNNs) on mobile devices, farmers can now identify diseases with 90% accuracy directly in the orchard. Advanced tools like thermal imaging and biosensors act as an early-warning system, detecting hidden stress before symptoms ever appear. This shift toward predictive farming empowers growers to protect their livelihoods more sustainably. Techniques like transfer learning bridge the gap between high-tech science and daily labour, ensuring that even with limited data, farmers have a dependable companion in the grove. Ultimately, integrating ML and internet of things (IoT) minimises chemical reliance, safeguarding both the environment and the long-term health of the durian industry. © The Author(s).
Department of Plant Science, International Islamic University Malaysia, Kuantan, 25200, Malaysia; Department of Biology Education, Universitas Negeri Yogyakarta, Yogyakarta, 55281, Indonesia
Research at a Glance
Register to unlock
Topics & SDG Alignment
Register to unlock
Collaboration
Register to unlock
Author Profile (Selected)
Register to unlock
References Overview
Register to unlock
Journal & Source
Register to unlock
Metadata & Integrity
Register to unlock