Advertisement

Multimedia Tools and Applications

, Volume 78, Issue 3, pp 3613–3632 | Cite as

Image based fruit category classification by 13-layer deep convolutional neural network and data augmentation

  • Yu-Dong ZhangEmail author
  • Zhengchao Dong
  • Xianqing Chen
  • Wenjuan Jia
  • Sidan Du
  • Khan MuhammadEmail author
  • Shui-Hua WangEmail author
Article

Abstract

Fruit category identification is important in factories, supermarkets, and other fields. Current computer vision systems used handcrafted features, and did not get good results. In this study, our team designed a 13-layer convolutional neural network (CNN). Three types of data augmentation method was used: image rotation, Gamma correction, and noise injection. We also compared max pooling with average pooling. The stochastic gradient descent with momentum was used to train the CNN with minibatch size of 128. The overall accuracy of our method is 94.94%, at least 5 percentage points higher than state-of-the-art approaches. We validated this 13-layer is the optimal structure. The GPU can achieve a 177× acceleration on training data, and a 175× acceleration on test data. We observed using data augmentation can increase the overall accuracy. Our method is effective in image-based fruit classification.

Keywords

Convolutional neural network Fully connected layer Softmax Fruit category identification 

Notes

Acknowledgments

This study was supported by Natural Science Foundation of China (61602250), Natural Science Foundation of Jiangsu Province (BK20150983), Open fund of Key Laboratory of Guangxi High Schools Complex System and Computational Intelligence (2016CSCI01).

Compliance with ethical standards

Conflict of interest

We have no conflicts of interest to disclose with regard to the subject matter of this paper.

References

  1. 1.
    Acquarelli J, van Laarhoven T, Gerretzen J et al (2017) Convolutional neural networks for vibrational spectroscopic data analysis. Anal Chim Acta 954:22–31CrossRefGoogle Scholar
  2. 2.
    Adak MF, Yumusak N (2016) Classification of E-Nose Aroma Data of Four Fruit Types by ABC-Based Neural Network. Sensors 16(3):13CrossRefGoogle Scholar
  3. 3.
    Ahmad J, Mehmood I, Baik SW (2017) Efficient object-based surveillance image search using spatial pooling of convolutional features. J Vis Commun Image Represent 45:62–76CrossRefGoogle Scholar
  4. 4.
    Bai X, Shi BG, Zhang CQ et al (2017) Text/non-text image classification in the wild with convolutional neural networks. Pattern Recogn 66:437–446CrossRefGoogle Scholar
  5. 5.
    Chen Y (2016) Voxelwise detection of cerebral microbleed in CADASIL patients by leaky rectified linear unit and early stopping: a class-imbalanced susceptibility-weighted imaging data study. Multime Tools Appl.  https://doi.org/10.1007/s11042-017-4383-9
  6. 6.
    Cicero M, Bilbily A, Dowdell T et al (2017) Training and Validating a Deep Convolutional Neural Network for Computer-Aided Detection and Classification of Abnormalities on Frontal Chest Radiographs. Investig Radiol 52(5):281–287CrossRefGoogle Scholar
  7. 7.
    Cintas C, Quinto-Sanchez M, Acuna V et al (2017) Automatic ear detection and feature extraction using Geometric Morphometrics and convolutional neural networks. IET Biometrics 6(3):211–223CrossRefGoogle Scholar
  8. 8.
    Dai-Ton H, Duc-Dung N, Duc-Hieu L (2016) An adaptive over-split and merge algorithm for page segmentation. Pattern Recogn Lett 80:137–143CrossRefGoogle Scholar
  9. 9.
    Deliens T, Deforche B, Annemans L et al (2016) Effectiveness of pricing strategies on french fries and fruit purchases among university students: results from an on-campus restaurant experiment. PLoS One 11(11):16 Article ID: e0165298CrossRefGoogle Scholar
  10. 10.
    Di Cagno R, Filannino P, Cavoski I et al (2017) Bioprocessing technology to exploit organic palm date (Phoenix dactylifera L. cultivar Siwi) fruit as a functional dietary supplement. J Funct Foods 31:9–19CrossRefGoogle Scholar
  11. 11.
    Garcia F, Cervantes J, Lopez A et al (2016) Fruit classification by extracting color chromaticity, shape and texture features: towards an application for supermarkets. IEEE Lat Am Trans 14(7):3434–3443CrossRefGoogle Scholar
  12. 12.
    Getahun S, Ambaw A, Delele M et al (2017) Analysis of airflow and heat transfer inside fruit packed refrigerated shipping container: Part I - model development and validation. J Food Eng 203:58–68CrossRefGoogle Scholar
  13. 13.
    Ghazi MM, Yanikoglu B, Aptoula E (2017) Plant identification using deep neural networks via optimization of transfer learning parameters. Neurocomputing 235:228–235CrossRefGoogle Scholar
  14. 14.
    Ji G (2014) Fruit classification using computer vision and feedforward neural network. J Food Eng 143:167–177CrossRefGoogle Scholar
  15. 15.
    Jiang YL, Zur RM, Pesce LL et al (2009) A Study of the Effect of Noise Injection on the Training of Artificial Neural Networks. In International Joint Conference on Neural Networks (IJCNN), IEEE, Atlanta, pp 2784–2788Google Scholar
  16. 16.
    Kim JH, Hong HG, Park KR (2017) Convolutional neural network-based human detection in nighttime images using visible light camera sensors. Sensors (Basel) 17(5).  https://doi.org/10.3390/s17051065
  17. 17.
    Kooi T, van Ginneken B, Karssemeijer N et al (2017) Discriminating solitary cysts from soft tissue lesions in mammography using a pretrained deep convolutional neural network. Med Phys 44(3):1017–1027CrossRefGoogle Scholar
  18. 18.
    Lee CH, Chien JT (2016) Deep unfolding inference for supervised topic model. In International Conference on Acoustics, Speech And Signal Processing Proceedings, IEEE, Shanghai, pp 2279–2283Google Scholar
  19. 19.
    Li S, Jiang H, Pang W (2017) Joint multiple fully connected convolutional neural network with extreme learning machine for hepatocellular carcinoma nuclei grading. Comput Biol Med 84:156–167CrossRefGoogle Scholar
  20. 20.
    Liu F, Snetkov L, Lima D (2017) Summary on fruit identification methods: A literature review. Adv Soc Sci Educ Hum Res 119:1629–1633Google Scholar
  21. 21.
    Lu Z (2016) Fractional Fourier entropy increases the recognition rate of fruit type detection. BMC Plant Biol 16(S2) Article ID: 10Google Scholar
  22. 22.
    Lu Z, Li Y (2017) A fruit sensing and classification system by fractional fourier entropy and improved hybrid genetic algorithm. In 5th International Conference on Industrial Application Engineering (IIAE). Kitakyushu, Institute of Industrial Applications Engineers, Japan, pp 293–299Google Scholar
  23. 23.
    Miki Y, Muramatsu C, Hayashi T et al (2017) Classification of teeth in cone-beam CT using deep convolutional neural network. Comput Biol Med 80:24–29CrossRefGoogle Scholar
  24. 24.
    Oliva D, Hinojosa S, Cuevas E et al (2017) Cross entropy based thresholding for magnetic resonance brain images using Crow Search Algorithm. Expert Syst Appl 79:164–180CrossRefGoogle Scholar
  25. 25.
    Pardo-Mates N, Vera A, Barbosa S et al (2017) Characterization, classification and authentication of fruit-based extracts by means of HPLC-UV chromatographic fingerprints, polyphenolic profiles and chemometric methods. Food Chem 221:29–38CrossRefGoogle Scholar
  26. 26.
    Qian RQ, Yue Y, Coenen F et al (2016) Traffic sign recognition with convolutional neural network based on max pooling positions. In 2th International Conference on Natural Computation, Fuzzy Systems And Knowledge Discovery (ICNC-FSKD), IEEE, Changsha pp 578–582Google Scholar
  27. 27.
    Radi, Ciptohadijoyo S, Litananda WS et al (2016) Electronic nose based on partition column integrated with gas sensor for fruit identification and classification. Comput Electron Agric 121:429–435CrossRefGoogle Scholar
  28. 28.
    Shao WH, Li YJ, Diao SF et al (2017) Rapid classification of Chinese quince (Chaenomeles speciosa Nakai) fruit provenance by near-infrared spectroscopy and multivariate calibration. Anal Bioanal Chem 409(1):115–120CrossRefGoogle Scholar
  29. 29.
    Sui XD, Zheng YJ, Wei BZ et al (2017) Choroid segmentation from Optical Coherence Tomography with graph edge weights learned from deep convolutional neural networks. Neurocomputing 237:332–341CrossRefGoogle Scholar
  30. 30.
    Tabik S, Peralta D, Herrera-Poyatos A et al (2017) A snapshot of image pre-processing for convolutional neural networks: case study of MNIST. Int J Comput Intellig Syst 10(1):555–568CrossRefGoogle Scholar
  31. 31.
    Teh V, Sim KS, Wong EK (2016) Brain early infarct detection using gamma correction extreme-level eliminating with weighting distribution. Scanning 38(6):842–856CrossRefGoogle Scholar
  32. 32.
    Thung KH, Paramesran R, Lim CL (2012) Content-based image quality metric using similarity measure of moment vectors. Pattern Recogn 45(6):2193–2204CrossRefGoogle Scholar
  33. 33.
    Thung KH, Wee CY, Yap PT et al (2014) Neurodegenerative disease diagnosis using incomplete multi-modality data via matrix shrinkage and completion. NeuroImage 91:386–400CrossRefGoogle Scholar
  34. 34.
    Thung KH, Wee CY, Yap PT et al (2016) Identification of progressive mild cognitive impairment patients using incomplete longitudinal MRI scans. Brain Struct Funct 221(8):3979–3995CrossRefGoogle Scholar
  35. 35.
    Tovar MF, Losada HV (2016) Fuzzy systems: case study classification of fruit Mc Stipitata Vaug (Araza). Amazonia Investiga 5(9):45–56Google Scholar
  36. 36.
    Wei L (2015) Fruit classification by wavelet-entropy and feedforward neural network trained by fitness-scaled chaotic ABC and biogeography-based optimization. Entropy 17(8):5711–5728Google Scholar
  37. 37.
    Wu L (2012) Classification of fruits using computer vision and a multiclass support vector machine. Sensors 12(9):12489–12505CrossRefGoogle Scholar
  38. 38.
    Wu J (2016) Fruit classification by biogeography-based optimization and feedforward neural network. Expert Syst 33(3):239–253CrossRefGoogle Scholar
  39. 39.
    Smirnov EA, Timoshenko DM, Andrianov SN (2014) Comparison of regularization methods for imagenet classification with deep convolutional neural networks. In 2nd Aasri Conference on Computational Intelligence And Bioinformatics (CIB). Elsevier Science Bv, South Korea, pp 89–94Google Scholar
  40. 40.
    Yaghoubi S, Noori S, Azaron A et al (2015) Resource allocation in multi-class dynamic PERT networks with finite capacity. Eur J Oper Res 247(3):879–894MathSciNetCrossRefGoogle Scholar
  41. 41.
    Zhang Y (2016) GroRec: a group-centric intelligent recommender system integrating social, mobile and big data technologies. IEEE Trans Serv Comput 9(5):786–795CrossRefGoogle Scholar
  42. 42.
    Zhang Y, Qiu M, Tsai CW et al (2015) Health-CPS: healthcare cyber-physical system assisted by cloud and big data. IEEE Syst J PP(99):1–8Google Scholar
  43. 43.
    Zhang Y, Chen M, Huang D et al (2017) iDoctor: Personalized and professionalized medical recommendations based on hybrid matrix factorization. Futur Gener Comput Syst 66:30–35CrossRefGoogle Scholar
  44. 44.
    Zhu SG, Du JP (2014) Visual tracking using max-average pooling and weight-selection strategy. J Appl Math 2014:828907Google Scholar

Copyright information

© Springer Science+Business Media, LLC 2017

Authors and Affiliations

  1. 1.School of Computer Science and TechnologyHenan Polytechnic UniversityJiaozuoPeople’s Republic of China
  2. 2.Jiangsu Key Laboratory of Advanced Manufacturing TechnologyHuaiyinChina
  3. 3.Translational Imaging Division & MRI UnitColumbia University and New York State Psychiatric InstituteNew YorkUSA
  4. 4.Department of electrical engineering, College of engineeringZhejiang Normal UniversityZhejiangChina
  5. 5.School of Computer Science and TechnologyNanjing Normal UniversityNanjingChina
  6. 6.School of Electronic Science and EngineeringNanjing UniversityNanjingChina
  7. 7.College of Software ConvergenceSejong UniversitySeoulRepublic of Korea

Personalised recommendations