[1]费琦琪,施杰,夏敏,等.基于CNN的冰糖橙分级系统[J].江苏农业学报,2020,(02):513-519.[doi:doi:10.3969/j.issn.1000-4440.2020.02.036]
　FEI Qi-qi,SHI Jie,XIA Min,et al.Grading system of Bingtang sweet orange based on convolutional neural networks[J].,2020,(02):513-519.[doi:doi:10.3969/j.issn.1000-4440.2020.02.036]
点击复制

基于CNN的冰糖橙分级系统()
分享到：

江苏农业学报[ISSN:1006-6977/CN:61-1281/TN]

卷:
期数:
2020年02期
页码:
513-519
栏目:
加工贮藏·质量安全
出版日期:
2020-04-30

文章信息/Info

Title:
Grading system of Bingtang sweet orange based on convolutional neural networks
作者:
费琦琪1; 施杰1; 夏敏1; 李刚2; 果霖1; 张天会1
（1.云南农业大学机电工程学院，云南昆明650201；2.昆明理工大学，云南昆明650504）
Author(s):
FEI Qi-qi1; SHI Jie1; XIA Min1; LI Gang2; GUO Lin1; ZHANG Tian-hui1
（1.College of Mechanical and Electrical Engineering, Yunnan Agricultural University, Kunming 650201, China；2.Kunming University of Science and Technology, Kunming 650504, China）
关键词:
冰糖橙分级; 积神经网络; LabVIEW; 系统设计
Keywords:
Bingtang sweet orange classification; convolutional neural networks(CNN); LabVIEW; system design
分类号:
TP274+3;S666.4
DOI:
doi:10.3969/j.issn.1000-4440.2020.02.036
文献标志码:
A
摘要:
为了提高冰糖橙的产业竞争力和效益，在售前对其进行分级是一道重要的工序。针对传统的冰糖橙表面缺陷分级方法存在工作繁琐且受人为因素干扰大的问题，本研究设计了一种将卷积神经网络（Convolutional Neural Networks，CNN）算法与虚拟仪器技术相结合的冰糖橙表面缺陷智能分级系统，并基于LabVIEW2018平台设计开发了一套冰糖橙分级系统。通过实验验证，该系统识别率达96.67%，验证了该分级方法和分级系统的有效性和可行性。
Abstract:
In order to improve the industrial competitiveness and efficiency of Bingtang sweet orange, grading is an important process before sale. In view of the traditional surface defect classification method of Bingtang sweet orange, there are some problems that the feature extraction is complicated and disturbed by human factors. In this study, an intelligent grading system for Bingtang sweet orange surface defect classification was designed based on convolutional neural networks(CNN) algorithm and virtual instrument technology. In addition, the Bingtang sweet orange grading system was designed and developed based on LabVIEW 2018. Through experimental verification, the recognition rate of the system reaches 96.67%, which proves the feasibility and effectiveness of the classification method and system.

参考文献/References:

[1]李梅.水果分拣技术的研究现状与发展[J].江苏理工学院学报,2018,24(2):121-124.
[2]AMIN N,AMIN T G,ZHANG Y D. Image-based deep learning automated sorting of date fruit[J]. Postharvest Biology and Technology,2019,153：133-141.
[3]SAJAD S,YOUSEF A G,GINS G M. A new approach for visual identification of orange varieties using neural networks and metaheuristic algorithms[J]. Information Processing in Agriculture,2018,5(1):162-172.
[4]王海青,姬长英,顾宝兴，等.基于机器视觉和支持向量机的温室黄瓜识别[J].农业机械学报,2012,43(3):163-167，180.
[5]赵娟,彭彦昆,SAGAR DHAKAL，等.基于机器视觉的苹果外观缺陷在线检测[J].农业机械学报,2013,44(S1):260-263.
[6]胡发焕,董增文,匡以顺.基于机器视觉的脐橙品质在线分级检测系统[J].中国农业大学学报,2016,21(3):112-118.
[7]MEGHA P A, LAKSHMANA. Computer vision based fruit grading system for quality evaluation of tomato in agriculture industry[J]. Procedia Computer Science,2016,79:426-433.
[8]卢勇威. 基于机器视觉的水果分拣系统[J].装备制造技术,2017(3):163-165，168.
[9]邓立苗,韩仲志,徐艳，等. 基于机器视觉的马铃薯智能分级系统[J].食品与机械,2014,30(5):144-146.
[10]HUBEL D H,WIESEL T N. Receptive fields, binocular interaction and functional architecture in the cat’s visual cortex[J]. The Journal of Physiology, 1962, 160(1):106-154.
[11]HINTON G E,OSINDERO S,THE Y W. A fast learning algorithm for deep belief nets[J]. Neural Computation, 2006, 18 (7): 1527-1554.
[12]郑远攀,李广阳,李晔. 深度学习在图像识别中的应用研究综述[J].计算机工程与应用,2019,55(12):20-36.
[13]施杰,李欢,果霖，等. 基于EMD和包络解调的轴承故障诊断系统研究[J].煤矿机械,2015,36(6):309-312.
[14]孙小明. 基于LabVIEW和Matlab混合编程的滚动轴承故障诊断系统[J].电子科技,2018,31(7):11-14.
[15]伍锡如,雪刚刚,刘英璇. 基于深度学习的水果采摘机器人视觉识别系统设计[J].农机化研究,2020,42(2):177-182，188.
[16]SMITH L N. Cyclical learning rates for training neural networks[C]//IEEE. 2017 IEEE Winter Conference on Applications of Computer Vision (WACV). Santa Rosa,USA：IEEE, 2017:464-472.
[17]赵建敏，李艳，李琦，等.基于卷积神经网络的马铃薯叶片病害识别系统[J].江苏农业科学，2018，46(24):251-255.

备注/Memo

备注/Memo:
收稿日期：2019-08-27基金项目：云南省重大科技专项(2018ZC001-303)；云南农业大学自然科学青年科研基金项目(2015ZR13)作者简介：费琦琪（1994-），女，安徽合肥人，硕士研究生，研究方向为机械制造及其自动化。（E-mail）1452931639@qq.com通讯作者：果霖,（E-mail）1265120@qq.com
更新日期/Last Update: 2020-05-18