摘要
深度学习作为图像识别领域重要的技术手段,具有识别速度快、准确率高等优势。阐明了深度学习技术研究的意义及必要性,概述了国内外深度学习领域农作物病虫害图像识别技术的研究进展,对深度学习技术在图像识别研究中存在的问题进行归纳总结,并指出深度学习领域中的图像识别方法存在训练样本大、模型结构复杂、复杂图像识别正确率低等问题。提出了一种CNN与胶囊网络的组合模型,经过初步实验,模型的图像识别正确率达93.75%,比CNN模型提高了3.55个百分点。随着深度学习技术的不断发展,胶囊网络研究将是未来的发展趋势。
Throughout the history of agricultural development,crop pests and diseases have always been one of the main obstacles hindering the development of agricultural economy.The crop disease identification system based on digital image processing technology had the characteristics of fast,accurate and real-time,which can help the farmers to take effective prevention measures in time.As an important technical means in the field of image recognition,deep learning has broad application prospects.The research progress of crop pest and disease image recognition technology in deep learning field in China and abroad was reviewed.The significance and necessity of deep learning technology research were clarified.The training samples of deep learning technology in image recognition research were large and the model structure was complex.Complex image recognition accuracy was low.It was proposed that improving the recognition accuracy of complex images would be the development direction of future research.
作者
贾少鹏
高红菊
杭潇
JIA Shaopeng;GAO Hongju;HANG Xiao(College of Information and Electrical Engineering,China Agricultural University,Beijing 100083,China)
出处
《农业机械学报》
EI
CAS
CSCD
北大核心
2019年第B07期313-317,共5页
Transactions of the Chinese Society for Agricultural Machinery
基金
国家自然科学基金项目(31371531)
关键词
病虫害
深度学习
图像识别
pests and diseases
deep learning
image recognition
