食品与发酵工业 >

2022 , Vol. 48 >Issue 18: 75 - 80

DOI: https://doi.org/10.13995/j.cnki.11-1802/ts.029675

研究报告

基于机器视觉的生鲜牛肉冷藏时间识别研究

  • 张茹 ,
  • 张奋楠 ,
  • 周星宇 ,
  • 俞经虎
展开
  • 1(江南大学 机械工程学院,江苏 无锡,214122)
    2(江苏省食品先进制造装备技术重点实验室,江苏 无锡,214122)
第一作者:硕士研究生(俞经虎教授为通信作者,E-mail:jhyu@jiangnan.edu.cn)

收稿日期: 2021-10-16

  修回日期: 2021-11-26

  网络出版日期: 2022-10-17

基金资助

江苏省先进食品制造装备与技术重点实验室项目(FMZ201901);国家自然科学基金项目(51375209)

收起

Identification of the refrigerated time for chilled beef based on machine vision

  • ZHANG Ru ,
  • ZHANG Fennan ,
  • ZHOU Xingyu ,
  • YU Jinghu
Expand
  • 1(School of Mechanical Engineering, Jiangnan University, Wuxi 214122, China)
    2(Jiangsu Province Key Laboratory of Advanced Food Manufacturing Equipment and Technology, Wuxi 214122, China)

Received date: 2021-10-16

  Revised date: 2021-11-26

  Online published: 2022-10-17

Fold

摘要

为解决生鲜牛肉冷藏时间人为识别误差大且无法量化的问题,该研究以4种冷藏时间共400块切片生鲜牛肉为研究对象,通过图像采集系统获取样本图像后搭建基于GoogLeNet的改进模型进行生鲜牛肉冷藏时间识别,并在此基础上引入迁移学习理论和数据增强技术来提高目标识别准确率。结果显示,该模型对测试集图像数据识别准确率为91.18%,迁移学习机制辅以数据增强技术能有效缓解复杂网络模型过拟合问题,并且在相同环境下的对比实验中,对生鲜牛肉冷藏时间的识别准确率优于反向传播式神经网络、深度卷积网络及GoogLeNet原始模型,可实现高效率、高准确性的生鲜牛肉冷藏时间识别,为生鲜牛肉的运输贮存安排及环境设置提供参考。

关键词： 机器视觉; 卷积神经网络; 迁移学习; 数据增强技术; 牛肉检测

本文引用格式

张茹 , 张奋楠 , 周星宇 , 俞经虎 . 基于机器视觉的生鲜牛肉冷藏时间识别研究[J]. 食品与发酵工业, 2022 , 48(18) : 75 -80 . DOI: 10.13995/j.cnki.11-1802/ts.029675

Abstract

To address the problem of the artificial recognition error and quantify the chilled beef refrigerated time the time, this study took 400 slices of chilled beef as raw materials with four kinds of refrigerated time to explore it. After obtaining the sample images through the image acquisition system, an improved model based on GoogLeNet was set up to identify the refrigerated time. Moreover, the transfer learning theory and data enhancement technology were applied to improve the accuracy of target recognition. The results showed that the recognition accuracy of the model was 91.18%, and the transfer learning mechanism with data enhancement technology could effectively alleviate the over-fitting problem of the complex network model. The results of the comparative experiments showed that the recognition accuracy of the method in this study was better than that of the back propagation neural network, the VGG convolution neural network and the GoogLeNet. In summary, the method in this research could realize the recognition of chilled beef refrigerated time with high efficiency and satisfactory accuracy. It could also provide a reference for the transportation schedule and environmental arrangement of chilled beef in the near future.

Key words： machine vision; convolutional neural network; transfer learning; data enhancement technology; beef detection

参考文献

[1] ALFAIA C M M, ALVES, S P, LOPES A F, et al.Effect of cooking methods on fatty acids, conjugated isomers of linoleic acid and nutritional quality of beef intramuscular fat[J].Meat Science, 2010, 84(4):769-777.
[2] 杜小粉, 董全.冷却肉的保鲜方法及其安全性研究进展[J].肉类研究, 2009, 23(4):35-38.
DU X F, DONG Q.Preservation of chilled meat and its safety[J].Meat Research, 2009, 23(4):35-38.
[3] LOWE D G.Object recognition from local scale -invariant features[C]//Proceedings of the 7th IEEE International Conference on Computer Vision.IEEE:1 150-1 157.
[4] DALAL N, TRIGGS B.Histograms of oriented gradients for human detection[C]//IEEE Computer Society Conference on Computer Vision and Pattern Recognition(CVPR 2005).IEEE:886-893.
[5] HOASHI H, JOUTOU T, YANAI K.Image recognition of 85 food categories by feature fusion[C]//IEEE International Symposium on Multimedia.IEEE:296-301.
[6] 马佳佳, 陈友鹏, 王克强, 等. 基于优化SVM的虫害图像识别研究[J].中国粮油学报, 2022, 37(5): 10-15.
MA J J, CHEN Y P, WANG K Q, et al. Research on insect pest image recognition based on optimized SVM[J]. Journal of the Chinese Cereals and Oils Association, 2022, 37(5): 10-15.
[7] 张继威, 牛少彰, 曹志义, 等.基于深度学习和约束稀疏表达的人脸识别算法[J].北京理工大学学报, 2019, 39(3):255-261.
ZHANG J W, NIU S Z, CAO Z Y,et al.Face recognition via deep learning and constraint sparse representation[J].Transactions of Beijing Institute of Technology, 2019, 39(3):255-261.
[8] 穆玉芝. 基于深度学习的语音信号处理研究[D].青岛:山东科技大学, 2019.
MU Y Z.Research on speech signal processing based on deep learning[D].Qingdao:Shandong University of Science and Technology, 2019.
[9] 乐万德, 任静, 刘舟洲, 等.精品西红柿尺寸分级检测算法[J].包装工程, 2021, 42(17):218-225.
YUE W D, REN J, LIU Z Z, et al.Size grading detection algorithm of high-quality tomatoes[J].Packaging Engineering, 2021, 42(17):218-225.
[10] 肖旺, 杨煜俊, 申启访, 等.基于改进的GoogLeNet鸭蛋表面缺陷检测[J].食品与机械, 2021, 37(6):162-167.
XIAO W, YANG Y J, SHEN Q F, et al.Duck egg surface defect detection based on improved GoogLeNet[J].Food and Machinery, 2021, 37(6):162-167.
[11] 刘运韬, 李渊, 刘逊韵. 基于可重叠混淆树的卷积神经网络[J]. 计算机应用研究, 2022, 39(3): 938-942.
LIU Y T, LI Y, LIU X Y. CT-CNN: Overlapping confusion tree based convolutional neural network[J]. Application Research of Computers, 2022, 39(3): 938-942.
[12] 王一丁, 姚毅, 李耀利, 等. 改进EfficientNet的表皮细胞图像识别研究[J].计算机工程与应用, 2022, 58(11): 200-208.
WANG Y D, YAO Y, LI Y L, et al. Improved efficientnet′s epidermal cell image recognition research[J]. Computer Engineering and Applications, 2022, 58(11): 200-208.
[13] 徐文龙. 基于卷积神经网络的红枣外观品质分级方法研究[D].天津:天津工业大学, 2019.
XU W L.Research on Jujube appearance quality classification method based on convolutional neural network[D].Tianjin:Tianjin Polytechnic University, 2019.
[14] 薛勇, 王立扬, 张瑜, 等.基于GoogLeNet深度迁移学习的苹果缺陷检测方法[J].农业机械学报, 2020, 51(7):30-35.
XUE Y, WANG L Y, ZHANG Y, et al.Defect detection method of apples based on Goog Le Net deep transfer learning[J].Transactions of the Chinese Society for Agricultural Machinery, 2020, 51(7):30-35.
[15] CHEN M, DHINGRA K, WU W, et al.PFID:Pittsburgh fast-food image dataset[C]//2009 16th IEEE International Conference on Image Processing (ICIP).IEEE:289-292.
[16] BOSSARD L, GUILLAUMIN M, GOOL L. Food-101-mining discriminative components with random forests[C]//Computer Vision-ECCV 2014, 2014.
[17] 廖恩红, 李会芳, 王华, 等.基于卷积神经网络的食品图像识别[J].华南师范大学学报(自然科学版), 2019, 51(4):113-119.
LIAO E H, LI H F, WANG H, et al.Food Image recognition based on convolutional neural network[J].Journal of South China Normal University(Natural Science Edition), 2019, 51(4):113-119.
[18] 孟一, 张玉华, 姜沛宏, 等.不同新鲜度牛肉特征性挥发成分的研究[J].现代食品科技, 2017, 33(4):271-277.
MENG Y, ZHANG Y H, JIANG P H, et al.Study of the characteristic volatile components in beef at different degrees of freshness[J].Modern Food Science and Technology, 2017, 33(4):271-277.
[19] 殷凯. 基于改进的Faster R-CNN图像人物检测研究[D].淮南:安徽理工大学, 2020.
YIN K.Research on image character detection based on improved faster R-CNN[D].Huainan:Anhui University of Science and Technology, 2020.
[20] XIE S P, ZHENG X Y, CHEN Y, et al.Artifact removal using improved GoogLeNet for sparse-view CT reconstruction[J].Scientific Reports, 2018, 8:6700.
[21] 徐一鸣, 张娟, 刘成成, 等.迁移学习模式下基于GoogLeNet网络的风电机组视觉检测[J].计算机科学, 2019, 46(5):260-265.
XU Y M, ZHANG J, LIU C C, et al.Wind turbine visual inspection based on GoogLeNet network in transfer learning mode[J].Computer Science, 2019, 46(5):260-265.
[22] 黄家才, 舒奇, 朱晓春, 等.基于迁移学习的机器人视觉识别与分拣策略[J].计算机工程与应用, 2019, 55(8):232-237.
HUANG J C, SHU Q, ZHU X C, et al.Robot vision recognition and sorting strategy based on transfer learning[J].Computer Engineering and Applications, 2019, 55(8):232-237.
[23] 许景辉, 邵明烨, 王一琛, 等.基于迁移学习的卷积神经网络玉米病害图像识别[J].农业机械学报, 2020, 51(2):230-236;253.
XU J H, SHAO M Y, WANG Y C, et al.Recognition of corn leaf spot and rust based on transfer learning with convolutional neural network[J].Transactions of the Chinese Society for Agricultural Machinery, 2020, 51(2):230-236;253.
Options
文章导航

/

〈 〉

技术支持：北京玛格泰克科技发展有限公司