Related Experiment Video
Updated: May 14, 2026
08:47
Computer Vision-Based Biomass Estimation for Invasive Plants
Published on: February 9, 2024
A Multimodal UAV-IoT Sensing Framework for Intelligent Pest Density Estimation in Smart Agricultural Systems
Yida Zhang1, Jianxi Chen2,3, Xin Zeng1
1China Agricultural University, Beijing 100083, China.
Sensors (Basel, Switzerland)
|May 13, 2026
Summary
This study introduces a multimodal pest density estimation framework (PDEF) using AI to integrate drone imagery, trap data, and environmental sensors for accurate agricultural pest monitoring. The AI-driven approach significantly improves pest prediction and crop management.
Area of Science:
- Agricultural Science
- Artificial Intelligence
- Environmental Monitoring
Background:
- Modern agriculture increasingly relies on intelligent sensing and AI for monitoring, but traditional single-source data approaches limit capturing complex plant-environment interactions.
- Existing methods struggle to integrate diverse data types for comprehensive pest analysis, hindering accurate decision-making in dynamic field conditions.
Purpose of the Study:
- To develop a multimodal pest density estimation framework (PDEF) that integrates UAV imagery, trap data, and environmental sensor measurements.
- To enhance the accuracy and applicability of pest monitoring models in complex agricultural environments by leveraging multi-source data fusion.
- To provide a novel AI-driven intelligent sensing framework for improved pest prediction and crop management.
Main Methods:
- Utilized convolutional neural networks (CNNs) for extracting crop canopy damage features from UAV imagery.
- Employed temporal encoding to model dynamic environmental variations and cross-modal feature alignment for deep multi-source information integration.
- Developed environment-aware enhancement mechanisms to create a unified feature representation space for improved estimation accuracy.
Main Results:
- The proposed Pest Density Estimation Framework (PDEF) achieved Mean Absolute Error (MAE) of 5.47, Root Mean Square Error (RMSE) of 7.62, and Mean Absolute Percentage Error (MAPE) of 14.9%.
- PDEF significantly outperformed a Transformer-based fusion model and demonstrated a coefficient of determination (R²) of 0.84, indicating superior fitting capability.
- Three-modality fusion reduced error metrics by over 20% compared to single-modality models, validating the effectiveness of multi-source collaborative modeling.
Conclusions:
- The multimodal pest density estimation framework (PDEF) offers a novel AI-driven approach for accurate pest monitoring by integrating diverse data sources.
- This framework enhances pest prediction capabilities and contributes to more intelligent agricultural production systems and data-driven decision-making.
- The study provides practical implications for agricultural economics and supply chain optimization through improved intelligent sensing systems.
Related Concept Videos
02:16Light Acquisition
In order to produce glucose, plants need to capture sufficient light energy. Many modern plants have evolved leaves specialized for light acquisition. Leaves can be only millimeters in width or tens of meters wide, depending on the environment. Due to competition for sunlight, evolution has driven the evolution of increasingly larger leaves and taller plants, to avoid shading by their neighbors with contaminant elaboration of root architecture and mechanisms to transport water and nutrients.
02:35Key Elements for Plant Nutrition
Like all living organisms, plants require organic and inorganic nutrients to survive, reproduce, grow and maintain homeostasis. To identify nutrients that are essential for plant functioning, researchers have leveraged a technique called hydroponics. In hydroponic culture systems, plants are grown—without soil—in water-based solutions containing nutrients. At least 17 nutrients have been identified as essential elements required by plants. Plants acquire these elements from the atmosphere, the...