基于红外热成像和机器学习的作物早期病害识别研究进展

Research progress of early crop disease identification based on infrared thermal imaging and machine learning

作物的早期病害检测作为针对性的防治手段已经成为智慧农业在病害方面的趋势,实现作物早期病害检测可以做到早发现,早治疗,减少作物农药使用,提高作物品质,减少经济损失。首先介绍作物病害的机制和红外热成像原理,发现红外热成像技术可以进行作物早期病害的检测;其次介绍红外热成像和机器学习的工作原理,综合概述国内外机器学习和红外热成像技术在病害识别领域的研究现状;分析红外热图像的缺点并使用机器学习进行改进,综述机器学习应用于处理红外热图像和红外热成像技术与机器学习相结合应用于作物病害的国内外现状,发现机器学习对红外热图像的缺点有着改进作用,还发现通过两种技术结合使用可以结合两者的优点进行更快更早的作物病害检测;最后通过分析现有研究成果,讨论现存的问题并提出相应的解决方法,对未来的研究趋势进行展望。

Early disease detection of crops as a targeted prevention and control method has become an important topic in smart agriculture.The ability to detect crop diseases at an early stage can significantly reduce the use of crop pesticides,improve crop quality,and reduce economic losses.This paper introduces the mechanism of crop diseases and the principle of infrared thermal imaging as the basis for early crop disease detection.It has been found that infrared thermal imaging technology can effectively detect early crop diseases.Next,the working principles of infrared thermal imaging and machine learning are introduced,and the research status of domestic and foreign machine learning and infrared thermal imaging of imaging technology in the field of disease identification is reviewed.Shortcomings of infrared thermal images are analyzed,and machine learning is proposed as a solution to improve the accuracy of early disease detection.The application of machine learning in processing infrared thermal images and the combination of infrared thermal imaging technology and machine learning for crop disease detection are summarized.Based on the current situation,it is found that machine learning can improve the shortcomings of infrared thermal images,and the advantages of the two technologies can be combined for faster and earlier detection of crop diseases.Finally,by analyzing existing research results,the paper proposes solutions to existing problems and discusses future research trends.