AI-Driven Pest Detection in Agriculture: Enhancing Accuracy and Processing Speed for Sustainable Crop Protection
P. Srilatha
Open and Distance Learning Centre (ODLC), Acharya N.G. Ranga Agricultural University (ANGRAU), Lam, Guntur, Andhra Pradesh, India.
Binseena S R *
Department of Entomology, Agriculture Research Station, Kerala Agricultural University, India.
Pritam Roy
Dairy Economics, Statistics and Management Division, ICAR-National Dairy Research Institute, Karnal, India.
Vikash Chandra Verma
Department of Agricultural Engineering, Bhola Paswan Shastri Agricultural College, Bihar Agricultural University, Sabour, Bhagalpur, India.
Rakesh Kumar
Punjab Agricultural University, Krishi Vigyan Kendra, Faridkot, India.
Ayushi Sharma
Division of Plant Pathology, SKUAST Jammu, Chatha, India.
Shabnam Kundal
Department of Agriculture Swami Sarvanand Institute of Management & Technology G.T. Road, Dinanagar- 143531, India.
Chandan Kumar Panigrahi
Department of Entomology, Faculty of Agricultural Sciences, Siksha 'O' Anusandhan Deemed to be University, Bhubaneswar - 751029 Odisha, India.
Deepika Sorahia
Department of Entomology, School of Crop Protection, College of Post Graduate Studies in Agricultural Sciences, Central Agricultural University (Imphal), Umiam, Meghalaya -793103, India.
Vimal Kumar
Department of Horticulture, University- IIMT University Meerut - 25001 (UP), India.
*Author to whom correspondence should be addressed.
Abstract
Pest detection is a critical component of effective crop management and agricultural productivity. Traditional methods of manual pest scouting are time-consuming and prone to human error. The advent of artificial intelligence algorithms has opened new avenues for improving pest detection accuracy and speed, thereby optimising the process. AI-based pest detection systems leverage machine learning techniques to analyse visual data captured by various sensing modalities, such as RGB cameras, multispectral imaging, and hyperspectral imaging. This review examines the current state of AI-based pest detection systems, focusing on deep learning architectures such as convolutional neural networks (CNNs) and object detection models. The challenges associated with developing robust pest detection algorithms include dataset quality, model generalisation, and real-time performance, which were discussed. Furthermore, the potential of integrating AI-based pest detection with precision agriculture techniques to enable targeted pest management interventions was discussed. Integrating multiple sensing modalities, such as RGB cameras, multispectral imaging, hyperspectral imaging, and acoustic sensors, can provide complementary information for pest detection. Techniques such as attention maps, feature visualisation, and rule extraction can provide insights into the learned features and decision-making process of pest detection models. Participatory sensing and citizen science can foster public awareness and engagement in pest management and support the co-creation of knowledge between researchers and stakeholders. The review concludes by outlining future research directions and the implications of AI-driven pest detection for sustainable agriculture.
Keywords: Pest detection, artificial intelligence, deep learning, convolutional neural networks, precision agriculture