深度学习在番茄叶片病害识别中的应用研究  

Research on the Application of Deep Learning in Tomato Leaf Disease Identification

作　　者：王新玉 刘坤[1] 李禹彤 魏传志 杜峰 WANG Xinyu;LIU Kun;LI Yutong;WEI Chuanzhi;DU Feng(College of Information and Electrical Engineering,Heilongjiang Bayi Agricultural University,Daqing,Heilongjiang 163319)

机构地区：[1]黑龙江八一农垦大学信息与电气工程学院,黑龙江大庆163319

出　　处：《热带农业工程》2025年第2期18-22,共5页Tropical Agricultural Engineering

基　　金：黑龙江省省属高等学校基本科研业务费科研项目(No.ZRCPY202021)。

摘　　要：近年来,随着深度学习技术的迅猛发展,番茄病害识别领域取得了显著进展。传统的番茄病害检测方法面临着效率低、检测精度不足和对环境变化适应性差等问题,而深度学习技术的应用为这一领域带来了突破性的改善。本文梳理了当前主流的深度学习模型及其在番茄病害识别中的应用,探讨深度学习在番茄叶片病害识别中的研究进展。研究表明,深度学习在番茄病害识别中应用效果显著,能够有效提高病害检测的准确性和效率。然而,该领域仍面临数据集规模不足、模型泛化能力差、实际应用环境复杂等问题。为解决这些问题,本文从数据集的扩充与标注质量提升、多模态数据融合、模型泛化能力的增强、迁移学习的应用、实时处理能力的提升、复杂背景下的识别研究等方面提出建议,将有助于进一步推动深度学习技术在番茄病害识别中的应用,促进农业数字化和智能化的发展。In recent years,with the rapid development of deep learning technology,significant progress has been made in the field of tomato disease identification.Traditional tomato disease detection methods face problems such as low efficiency,insufficient detection accuracy and poor adaptability to environmental changes,while the application of deep learning technology has brought breakthrough improvements in this field.In this paper,the current mainstream deep learning models and their applications in tomato disease identification are sorted out.The research progress of deep learning in tomato leaf disease recognition is discussed.The study shows that the application of deep learning in tomato disease recognition is effective and can effectively improve the accuracy and efficiency of disease detection.However,the field still faces problems such as insufficient dataset size,poor model generalization ability,and complex practical application environment.To solve these problems,this paper suggests that future research should focus on the following aspects:dataset expansion and annotation quality improvement,multimodal data fusion,enhancement of model generalization ability,application of transfer learning,improvement of real-time processing ability,and recognition research in complex contexts.These researches will help to further promote the application of deep learning technology in tomato disease recognition and promote the development of agricultural digitalization and intelligence.

关 键 词：深度学习 番茄叶片 病害识别 

分 类 号：TP23[自动化与计算机技术—检测技术与自动化装置]