Evaluation of State-of-the-Art Deep Learning Techniques for Plant Disease and Pest Detection

2025-08-09

AI Summary

To address the limitations of manual, low-accuracy, and inefficient plant disease and pest detection in traditional agriculture, this study systematically evaluates deep learning–based image recognition techniques. We propose a five-category methodological taxonomy encompassing hyperspectral imaging, explainable visualization, enhanced CNN architectures, Vision Transformers (ViTs), and a novel Hybrid vision Transformer (HvT). Evaluated on diverse multi-source agricultural image datasets, HvT achieves a state-of-the-art accuracy of 99.3%, substantially outperforming lightweight models such as MobileNetV3. Comprehensive benchmarking further confirms ViT-based models’ superior balance between detection accuracy and inference speed. The work delivers a practical, empirically grounded AI monitoring technology selection guide and performance benchmark for precision agriculture, bridging the gap between algorithmic validation and real-world deployment in intelligent crop protection.

Technology Category

• Computer Vision: Multi-modal Vision
• Machine Learning: Deep Neural Architectures and Foundation Models
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs

Abstract

Addressing plant diseases and pests is critical for enhancing crop production and preventing economic losses. Recent advances in artificial intelligence (AI), machine learning (ML), and deep learning (DL) have significantly improved the precision and efficiency of detection methods, surpassing the limitations of manual identification. This study reviews modern computer-based techniques for detecting plant diseases and pests from images, including recent AI developments. The methodologies are organized into five categories: hyperspectral imaging, non-visualization techniques, visualization approaches, modified deep learning architectures, and transformer models. This structured taxonomy provides researchers with detailed, actionable insights for selecting advanced state-of-the-art detection methods. A comprehensive survey of recent work and comparative studies demonstrates the consistent superiority of modern AI-based approaches, which often outperform older image analysis methods in speed and accuracy. In particular, vision transformers such as the Hierarchical Vision Transformer (HvT) have shown accuracy exceeding 99.3% in plant disease detection, outperforming architectures like MobileNetV3. The study concludes by discussing system design challenges, proposing solutions, and outlining promising directions for future research.

Problem

• Evaluating deep learning for plant disease and pest detection
• Comparing AI techniques to improve detection accuracy and speed
• Addressing challenges in system design for future research

Innovation

• Uses hyperspectral imaging for disease detection
• Employs modified deep learning architectures
• Leverages vision transformers like HvT