Context-Aware Pesticide Recommendation via Few-Shot Pest Recognition for Precision Agriculture

TL;DR

This work tackles the challenge of scalable, eco-friendly pest management for sugarcane and wheat by proposing a two-module framework that combines a lightweight Pest Detection Module built on a prototypical meta-learning-based few-shot CNN with a rule-based Pesticide Recommendation Module that incorporates crop type, growth stage, and pest severity. A comprehensive dataset for sugarcane and wheat pests is created by merging three public sources and resizing images to $100\\times100$, enabling effective few-shot learning with about $10$ million parameter parameters in the proposed Lightweight Prototype Network Backbone. The system demonstrates competitive pest detection accuracy against larger baselines across $1$- to $5$-shot scenarios and provides context-aware pesticide guidance, aiming to reduce chemical usage while enabling real-time deployment on edge devices like smartphones and drones. The combination of high-performance, low-complexity detection and actionable, eco-friendly recommendations offers a practical path toward sustainable precision agriculture and improved food security. Future work will broaden pest coverage, integrate real-time IoT environmental data, and extend the framework to additional crops and agricultural conditions.

Abstract

Effective pest management is crucial for enhancing agricultural productivity, especially for crops such as sugarcane and wheat that are highly vulnerable to pest infestations. Traditional pest management methods depend heavily on manual field inspections and the use of chemical pesticides. These approaches are often costly, time-consuming, labor-intensive, and can have a negative impact on the environment. To overcome these challenges, this study presents a lightweight framework for pest detection and pesticide recommendation, designed for low-resource devices such as smartphones and drones, making it suitable for use by small and marginal farmers. The proposed framework includes two main components. The first is a Pest Detection Module that uses a compact, lightweight convolutional neural network (CNN) combined with prototypical meta-learning to accurately identify pests even when only a few training samples are available. The second is a Pesticide Recommendation Module that incorporates environmental factors like crop type and growth stage to suggest safe and eco-friendly pesticide recommendations. To train and evaluate our framework, a comprehensive pest image dataset was developed by combining multiple publicly available datasets. The final dataset contains samples with different viewing angles, pest sizes, and background conditions to ensure strong generalization. Experimental results show that the proposed lightweight CNN achieves high accuracy, comparable to state-of-the-art models, while significantly reducing computational complexity. The Decision Support System additionally improves pest management by reducing dependence on traditional chemical pesticides and encouraging sustainable practices, demonstrating its potential for real-time applications in precision agriculture.

Figures (7)

• Figure 1: Overview of our proposed framework integrating a lightweight prototypical network for few-shot pest detection with the rule-based Decision Support System (integrates pest detection results with environmental factors) to recommend context-aware pesticides.
• Figure 2: Architecture of propose Lightweight Prototype Network Backbone.
• Figure 3: Architecture of FEBB
• Figure 4: Training loss of ResNet18.
• Figure 5: Training loss of DensNet169.