摘要
植物识别领域的研究包括单一背景和自然环境植物图像识别,由于背景噪声的存在,自然环境植物图像识别难度更大。针对如何降低卷积神经网络(CNN)的权重大小、如何改善过拟合、如何提高模型对自然环境植物的识别率和泛化能力的问题,提出科优先(FP)的植物识别方法。与轻量卷积神经网络Mobile Net模型结合,利用迁移学习的方法,建立基于Mobile Net的科优先(FP-MobileNet)植物识别模型。单纯使用Mobile Net模型在单一背景植物数据集flavia上获得了99.8%的识别率;对于更具挑战的自然环境花卉数据集flower102,在训练集样本数量大于测试集时FPMobile Net获得了99.56%识别率,在训练集样本数量小于测试集时FP-MobileNet仍获得了95.56%的识别率。实验结果表明,两种数据集划分方案下FP-MobileNet的识别率均高于单纯的Mobile Net模型;并且FP-MobileNet模型在获得较高识别率的同时,权重仅占13.7 MB,兼顾了精度和延迟,适合推广到需要轻量模型的移动设备。
Plant recognition includes two kinds of tasks:specimen recognition and real-environment recognition.Due to the existence of background noise,real-environment plant image recognition is more difficult.To reduce the weight of Convolutional Neural Networks(CNN),to improve over-fitting,to improve the recognition rate and generalization ability,a method of plant identification with Family Priority(FP)was proposed.Combined with the lightweight CNN MobileNet model,a plant recognition model Family Priority MobileNet(FP-MobileNet)was established by means of migration learning.On the single background plant dataset flavia,the MobileNet model achieved 99.8%of accuracy.For the more challenging real-environment flower dataset flower102,when the number of samples in the training set was greater than that in the test set FP-MobileNet achieved 99.56%of accuracy.When the number of samples in the training set was smaller than that in the test set,FP-MobileNet still obtained 95.56%of accuracy.The experimental results show that the accuracies of FP-MobileNet under two different data set partitioning schemes are both higher than those of the pure MobileNet model.In addition,FP-MobileNet weighs only occupy 13.7 MB with high recognition rate.It takes into account both accuracy and delay,and is suitable for promotion to mobile devices that require a lightweight model.
作者
曹香滢
孙卫民
朱悠翔
钱鑫
李晓宇
业宁
CAO Xiangying;SUN Weimin;ZHU Youxiang;QIAN Xin;LI Xiaoyu;YE Ning(School of Information Technology,Nanjing Forestry University,Nanjing Jiangsu 210037,China;Housing and Real Estate Promotion Center of Jiangsu Provincial Department of Housing and Urban Rural Development,Nanjing Jiangsu 210009,China)
出处
《计算机应用》
CSCD
北大核心
2018年第11期3241-3245,共5页
journal of Computer Applications
基金
国家重点研发计划项目(2016YFD0600101)
国家自然科学基金资助项目(31570662
31500533
61401214)
江苏省大学生创新计划项目(201710298029Z)
江苏省住建厅科技计划项目(2016ZD44)
江苏省优势学科项目~~
关键词
科优先策略
自然环境植物图像
植物图像识别
深度学习
卷积神经网络
Family Priority(FP)strategy
real-environment plant image
plant image recognition
deep learning
Convolutional Neural Network(CNN)
