Research Article | | Peer-Reviewed

Optimizing Food101 Classification with Transfer Learning: A Fine-Tuning Approach Using EfficientNetB0

Received: 1 July 2024     Accepted: 24 July 2024     Published: 15 August 2024
Views:       Downloads:
Abstract

Much research has been done on the classification of the food101 dataset, but much of this research which achieved an accuracy score of more than 90% explores heavyweight architecture such as EfficientNetB7, Visual Geometry Group19, ResNet-200, Inception v4, DenseNet-201, ResNeXt-101, MobileNet v3 and many more. This study explores the classification of the Food101 dataset using the EfficientNetB0 architecture, a lightweight architecture. Compared to other popular CNN architecture, EfficientNetB0 has relatively small parameters, which makes it computationally efficient and suitable for deployment on resource-constraint environments. The research aims to balance model accuracy and computational efficiency, addressing the need for resource-constrained environments. Five experiments were conducted while varying the number of fine-tuned layers. Results demonstrate that the fine-tuned EfficientNetB0 model achieves an accuracy score of accuracy score of 97.54%, Top_k_categorical accuracy of 99.89%, precision of 98.21%, and recall of 97.02% in just 5 epochs. This research will significantly contribute to the field of transfer learning by developing specialized models that excel in target tasks. Besides, it will advance dietary monitoring, food logging, and health-related technologies, enabling more accessible and practical solutions for consumers. However, the optimal number of layers to fine-tune for achieving perfect accuracy with EfficientNetB0 remains uncertain. It often involves trial and error to determine the best configuration for optimal results, presenting an opportunity for future research.

Published in International Journal of Intelligent Information Systems (Volume 13, Issue 4)
DOI 10.11648/j.ijiis.20241304.11
Page(s) 59-77
Creative Commons

This is an Open Access article, distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution and reproduction in any medium or format, provided the original work is properly cited.

Copyright

Copyright © The Author(s), 2024. Published by Science Publishing Group

Keywords

Transfer Learning, EfficientNets, Lightweight Architecture, Convolutional Neural Network, Fine-Tuning

References
[1] Keiron, S., & Ryan, N. (2015). An introduction to convolutional neural networks. arXiv, 1511.08458v2 [cs.NE].
[2] Shahid, N., Rappon, T., & Berta, W. (2019). Applications of artificial neural networks in health care organizational decision-making: A scoping review. PLoS One, 14(2), e0212356.
[3] Ciresan, D., Meier, U., & Schmidhuber, J. (2012). Multi-column deep neural networks for image classification. In Computer Vision and Pattern Recognition (CVPR), 2012 IEEE Conference on (pp. 3642–3649).
[4] Mader, K. (2018). Food 101 datasets. Kaggle.
[5] Bossard, L., Guillaumin, M., & Van Gool, L. (2014). Food-101 – Mining discriminative components with random forests. European Conference on Computer Vision.
[6] Ren, Z. T., Chen, X., & Wong, K. H. (2021). Neural architecture search for lightweight neural network in food recognition. Mathematics, 9(11), 1245.
[7] Mingxing T., Quoc, V. (2019). EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks. International Conference on Machine Learning, arXiv: 1905.11946 [cs.LG],
[8] Tan, M. (2019). EfficientNet: Improving accuracy and efficiency through AutoML and model scaling. Google Research Blog.
[9] Mathswork. (2021). EfficientNetB0. MathWorks.
[10] Sanchez, J., Perronnin, F., Mensink, T., & Verbeek, J. (2013). Image classification with the Fisher vector: Theory and practice. International Journal of Computer Vision, 105(3), 222-245.
[11] Lazebnik, S., Schmid, C., & Ponce, J. (2006). Beyond bags of features: Spatial pyramid matching for recognizing natural scene categories. In 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'06) (pp. 2169-2178). IEEE.
[12] Joutou, T., & Yanai, K. (2009). A food image recognition system with multiple kernel learning. In Proceedings of the 16th International Conference on Image Processing (pp. 285-288). IEEE.
[13] Chen, M. Y., et al. (2009). Automatic Chinese food identification and quantity estimation. SIGGRAPH Asia Technical Briefs.
[14] Chen, M., Dhingra, K., Wu, W., Yang, L., Sukthankar, R., & Yang, J. (2009). PFID: Pittsburgh fast- food image dataset. In ICIP.
[15] Hassannejad, H., Matrella, G., Ciampolini, P., De Munari, I., Mordonini, M., & Cagnoni, S. (2016). Food image recognition using very deep convolutional networks. Proceedings of the 2nd International Workshop on Multimedia Assisted Dietary Management (pp. 41–49). ACM.
[16] Lee, K. H., He, X., Zhang, L., & Yang, L. (2017). CleanNet: Transfer learning for scalable image classifier training with label noise. arXiv.
[17] Singh, P., & Susan, S. (2023). Transfer learning using very deep pre-trained models for food image classification. 2023 International Conference on Computing, Communication and Networking Technologies (ICCCNT).
[18] Rudraja, V. (2022). Food image classification using various CNN models. International Journal of Innovative Research in Technology, 9(3), 626.
[19] VijayaKumari, G., Priyanka, V., & Vishwanath, P. (2022). Food classification using transfer learning technique. Global Transitions Proceedings, 3(1), 225-229.
[20] Hosna, A., Merry, E., & Gyalmo, J. (2022). Transfer learning: A friendly introduction. Journal of Big Data, 9, 102.
[21] Jenan, A., A, & Raidah, S., K (2023). Integration of EfficientNetB0 and Machine Learning for Fingerprint Classification, Informatica, 49–56,
[22] Ahmed, T., & Sabab, N. H. (2020). Classification and understanding of cloud structures via satellite images with EfficientUNet. Earth and Space Science Open Archive.
[23] Wijdan, R., A., Nidhal, K., E., & Abdul, M., G. (2021). Hybrid Deep Neural Network for Facial Expressions Recognition. Indonesian Journal of Electrical Engineering and Informatics (IJEEI), 9(4), 993-1007, ISSN: 2089-3272,
[24] Neha, S., Sheifali, G., Mana, S. Reshan, A., Adel, S., Hani, A., Asadullah, S. (2021). EfficientNetB0 cum FPN Based Semantic Segmentation of Gastrointestinal Tract Organs in MRI Scans. National Centre of Biotechnology Information, 13(14): 2399.
[25] Paolo, D., A., Vito, P., P., Lorenzo, R., A., Francesca, O., B. (2024). Top-tuning: A study on transfer learning for an efficient alternative to fine tuning for image classification with fast kernel methods. Image and Vision Computing,
[26] Manoj, K., S., Brajesh, K. (2023). Fine tuning the pre-trained Convolutional Neural Network models for Hyperspectral Image Classification using transfer learning. Computer Vision and Robotics, 271-283,
[27] Jorge, S., Florent, P., Thomas, M., & Jakob, V. (2013). Image Classification with the Fisher Vector: Theory and Practice. International Journal of Computer Vision, 105(3),
[28] Taichi, J., & Keiji, Y. (2009). A food image recognition system with Multiple Kernel Learning. International Conference on Image Processing, 285 - 288,
[29] Mei-Yun, C., Yung-Hsiang, Y., Chia-Ju, H., & Shih-Han, W. (2012). Automatic Chinese food identification and quantity estimation. SIGGRAPH Asia 2012 Technical Briefs Conference,
[30] Lukas, B. Matthieu, G., & Luc-Van, G. (2014). Food-101 – mining discriminative components with Random Forests. Conference: European Conference on Computer Vision,
[31] Kuang-Huei, L., Xiaodong, H., Lei, Z., & Linjun, Y. (2018). Food-101N dataset.
[32] Francis, J., P., & Alon, S., A. (2021). Empirical analysis of a fine-tuned Deep Convolutional Model in classifying and detecting malaria parasites from blood smears. Transactions on Internet and Information Systems, 15(1): 147-165,
[33] Oguzhan, T., & Tahir, C. (2023). A review of transfer learning: Advantages, strategies and types. International Conference on Modern and Advanced Research.
[34] Tan, M. (2018). MnasNet: Towards automating the design of mobile machine learning models. Google Brain Team.
[35] Ahdi, M. W., Sjamsuri, K., Kunaefi, A., & Yusuf, A. (2023). Convolutional neural network (CNN) EfficientNet-B0 model architecture for paddy diseases classification. 14th International Conference on Information & Communication Technology and System (ICTS).
[36] Ghandour, C., El-Shafai, W., & El-Rabaie, S. (2023). Medical image enhancement algorithms using deep learning-based convolutional neural networks. Journal of Optics, 1-11.
[37] Yixing, F. (2020). Image classification via fine-tuning with EfficientNet.
[38] Venkatesh, B. (2021). How does the machine read images and use them in computer vision? Topcoder.
[39] Zhou, K., Oh, S. K., Pedrycz, W., & Qiu, J. (2023). Data preprocessing strategy in constructing convolutional neural network classifier based on constrained particle swarm optimization with fuzzy penalty function. Engineering Applications of Artificial Intelligence, 117, 105580.
[40] Yousif, M., & Balfaqih, M. (2023). Enhancing the accuracy of image classification using deep learning and preprocessing methods. Artificial Intelligence and Robotics Development Journal, 3(4), 269-281.
[41] Norhikmah, R., Lutfhi, A., & Rumini. (2022). The effect of layer batch normalization and dropout on CNN model performance for facial expression classification. International Journal on Informatics Visualization.
[42] Şengöz, N., Yiğit, T., Özmen, Ö., & Isik, A. H. (2022). Importance of preprocessing in histopathology image classification using deep convolutional neural networks. Advances in Artificial Intelligence Research, 2(1), 1-6.
[43] Pavlo, R. (2017). Impact of training set batch size on the performance of convolutional neural networks for diverse datasets. Information Technology and Management Science, 20(1), 20-24.
[44] Sakib, M., & Fang-Xiang, W. (2021). Diagnosis of autism spectrum disorder with convolutional autoencoder and structural MRI images. Neural Engineering Techniques for Autism Spectrum Disorder, 1(3), 23-38.
[45] Wang, et al. (2019). What is a convolutional neural network? CNN explainer.
[46] Klingler, N. (2024). EfficientNet: Optimizing deep learning efficiency. Viso.ai.
[47] Kattenborn, T., Leitloff, J., Schiefer, F., & Hinz, S. (2021). Review on convolutional neural networks (CNN) in vegetation remote sensing. ISPRS Journal of Photogrammetry and Remote Sensing, 173, 24-49.
[48] Toriba Scientific. (2023). Batch processing. HORIBA.
[49] Yogeshwari, M., & Thailambal, G. (2023). Automatic feature extraction and detection of plant leaf disease using GLCM features and convolutional neural networks. Materials Today: Proceedings, 81, 530-536.
[50] Pranjal, S., & Seba, S. (2023). Transfer learning using very deep pre-trained models for food image classification. 14th International Conference on Computing Communication and Networking Technologies (ICCCNT).
Cite This Article
  • APA Style

    Philip, A. R. (2024). Optimizing Food101 Classification with Transfer Learning: A Fine-Tuning Approach Using EfficientNetB0. International Journal of Intelligent Information Systems, 13(4), 59-77. https://doi.org/10.11648/j.ijiis.20241304.11

    Copy | Download

    ACS Style

    Philip, A. R. Optimizing Food101 Classification with Transfer Learning: A Fine-Tuning Approach Using EfficientNetB0. Int. J. Intell. Inf. Syst. 2024, 13(4), 59-77. doi: 10.11648/j.ijiis.20241304.11

    Copy | Download

    AMA Style

    Philip AR. Optimizing Food101 Classification with Transfer Learning: A Fine-Tuning Approach Using EfficientNetB0. Int J Intell Inf Syst. 2024;13(4):59-77. doi: 10.11648/j.ijiis.20241304.11

    Copy | Download

  • @article{10.11648/j.ijiis.20241304.11,
      author = {Adebayo Rotimi Philip},
      title = {Optimizing Food101 Classification with Transfer Learning: A Fine-Tuning Approach Using EfficientNetB0
    },
      journal = {International Journal of Intelligent Information Systems},
      volume = {13},
      number = {4},
      pages = {59-77},
      doi = {10.11648/j.ijiis.20241304.11},
      url = {https://doi.org/10.11648/j.ijiis.20241304.11},
      eprint = {https://article.sciencepublishinggroup.com/pdf/10.11648.j.ijiis.20241304.11},
      abstract = {Much research has been done on the classification of the food101 dataset, but much of this research which achieved an accuracy score of more than 90% explores heavyweight architecture such as EfficientNetB7, Visual Geometry Group19, ResNet-200, Inception v4, DenseNet-201, ResNeXt-101, MobileNet v3 and many more. This study explores the classification of the Food101 dataset using the EfficientNetB0 architecture, a lightweight architecture. Compared to other popular CNN architecture, EfficientNetB0 has relatively small parameters, which makes it computationally efficient and suitable for deployment on resource-constraint environments. The research aims to balance model accuracy and computational efficiency, addressing the need for resource-constrained environments. Five experiments were conducted while varying the number of fine-tuned layers. Results demonstrate that the fine-tuned EfficientNetB0 model achieves an accuracy score of accuracy score of 97.54%, Top_k_categorical accuracy of 99.89%, precision of 98.21%, and recall of 97.02% in just 5 epochs. This research will significantly contribute to the field of transfer learning by developing specialized models that excel in target tasks. Besides, it will advance dietary monitoring, food logging, and health-related technologies, enabling more accessible and practical solutions for consumers. However, the optimal number of layers to fine-tune for achieving perfect accuracy with EfficientNetB0 remains uncertain. It often involves trial and error to determine the best configuration for optimal results, presenting an opportunity for future research.
    },
     year = {2024}
    }
    

    Copy | Download

  • TY  - JOUR
    T1  - Optimizing Food101 Classification with Transfer Learning: A Fine-Tuning Approach Using EfficientNetB0
    
    AU  - Adebayo Rotimi Philip
    Y1  - 2024/08/15
    PY  - 2024
    N1  - https://doi.org/10.11648/j.ijiis.20241304.11
    DO  - 10.11648/j.ijiis.20241304.11
    T2  - International Journal of Intelligent Information Systems
    JF  - International Journal of Intelligent Information Systems
    JO  - International Journal of Intelligent Information Systems
    SP  - 59
    EP  - 77
    PB  - Science Publishing Group
    SN  - 2328-7683
    UR  - https://doi.org/10.11648/j.ijiis.20241304.11
    AB  - Much research has been done on the classification of the food101 dataset, but much of this research which achieved an accuracy score of more than 90% explores heavyweight architecture such as EfficientNetB7, Visual Geometry Group19, ResNet-200, Inception v4, DenseNet-201, ResNeXt-101, MobileNet v3 and many more. This study explores the classification of the Food101 dataset using the EfficientNetB0 architecture, a lightweight architecture. Compared to other popular CNN architecture, EfficientNetB0 has relatively small parameters, which makes it computationally efficient and suitable for deployment on resource-constraint environments. The research aims to balance model accuracy and computational efficiency, addressing the need for resource-constrained environments. Five experiments were conducted while varying the number of fine-tuned layers. Results demonstrate that the fine-tuned EfficientNetB0 model achieves an accuracy score of accuracy score of 97.54%, Top_k_categorical accuracy of 99.89%, precision of 98.21%, and recall of 97.02% in just 5 epochs. This research will significantly contribute to the field of transfer learning by developing specialized models that excel in target tasks. Besides, it will advance dietary monitoring, food logging, and health-related technologies, enabling more accessible and practical solutions for consumers. However, the optimal number of layers to fine-tune for achieving perfect accuracy with EfficientNetB0 remains uncertain. It often involves trial and error to determine the best configuration for optimal results, presenting an opportunity for future research.
    
    VL  - 13
    IS  - 4
    ER  - 

    Copy | Download

Author Information
  • Sections