Design and Implementation of Crop Leaf Disease Detection System Based on Deep Learning

Abstract

The precise and rapid detection of crop leaf diseases is a critical component in ensuring stable agricultural production and income growth, as well as advancing the development of smart agriculture. Traditional disease detection methods rely on manual observation and laboratory analysis, which suffer from low efficiency, subjective influence on identification results, and difficulties in adapting to large-scale field operations. Deep learning-based object detection algorithms, with their advantages of automated feature extraction and high recognition accuracy, provide a novel solution for crop leaf disease detection. This paper focuses on the YOLOv8 algorithm to construct a crop leaf disease recognition model, designing and implementing a disease detection system that integrates user management, multi-format image recognition, and flexible model switching. The system adheres to modular and hierarchical architectural design principles, utilizes the PyQt framework to develop a user-friendly interface, and employs SQLite database for efficient storage and management of user information and detection records. Tests demonstrate that the system features simple operation, rapid response, and accurate identification results, effectively enhancing the efficiency of crop leaf disease detection. It provides intelligent technical support for disease prevention and control in agricultural production, demonstrating strong practical application value.

Plant phenotyping relevance

作物葉の病害状態を画像から推定する深層学習モデルと検出システムの設計・実装が中心であり、植物病害フェノタイプの取得手法に該当する。

Code and data availability

公開本文の所在を確認できませんでした。非公開または購読が必要な可能性があります。

No evidence-backed public reproduction asset is currently recorded.