摘要
针对传统基于内容的识别方法在特征提取方面存在计算复杂、特征不可迁移等问题,为避免光照条件、重叠及其他遮挡等因素对图像识别的影响,以LeNet卷积神经网络作为基础,对其结构进行改进,设计了一种基于改进LeNet卷积神经网络的苹果目标识别模型,并利用该模型对不同场景的苹果图像进行识别训练与验证。结果表明:该网络模型可有效实现苹果图像的识别,对独立果实、遮挡果实、重叠果实以及相邻果实的识别率分别为96.25%,91.37%,94.91%,89.56%,综合识别率达到93.79%。与其他方法相比,该算法具有较强的抗干扰能力,图像识别速度快、识别率更高。
In order to avoid the influence of illumination condition,overlap and other occlusion on image recognition,an improved LeNet convolution neural network is used to improve the structure of the traditional content-based recognition method.An Apple target recognition model based on the improved LeNet convolution neural network is designed and used to avoid the influence of illumination condition,overlap and other occlusion factors on image recognition.The model trains and validates Apple images in different scenarios.The results show that the network model can effectively recognize apple images.The recognition rates of independent fruits,occluded fruits,overlapping fruits and adjacent fruits are 96.25%,91.37%,94.91% and 89.56%respectively,and the comprehensive recognition rate is 93.79%.Compared with other methods,this algorithm has stronger antijamming ability,faster image recognition speed and higher recognition rate.
作者
程鸿芳
张春友
CHENG Hong-fang;ZHANG Chun-you(Widiu Institute of Technology,Wuhu,Anhui 241000,China;College of Mechanical Engineering,Inner Mongolia University for the Nationalities,Tongliao,Inner Mongolia 028043,China)
出处
《食品与机械》
北大核心
2019年第3期155-158,共4页
Food and Machinery
基金
2017年安徽省高校科学研究项目(编号:KJ2017A560)
2017年高校优秀青年人才支持计划项目(编号:gxyqZD2017141)
2016年安徽省质量工程项目(编号:2016ckjh224)
芜湖职业技术学院科技创新团队(编号:Wzykj2018A02)
关键词
图像识别
目标识别
卷积神经网络
LeNet
image recognition
target recognition
convolution neural network
LeNet
