Journal of Applied Science and Engineering

Published by Tamkang University Press

ESCI jase impact factor scopus logo open access rate of Scopus journal

Automatic recognition and classification of product packaging images based on convolutional neural network

 2026

 Volume 32

You Li1 and Chen Jiang2

1College of Art and Design, Wuhan Technology and Business University, Wuhan, Hubei 430065, China

2School of art and media, Wuhan College, Wuhan, Hubei 430065, China

Received: February 12, 2026
Accepted: April 4, 2026
Publication Date: May 17, 2026

上傳圖片

Overall architecture of MSAFNet. The framework integrates a hybrid backbone network, attention modules, and a multi-scale feature pyramid to improve feature extraction and object detection performance for product packaging images. 

 Copyright The Author(s). This is an open access article distributed under the terms of the Creative Commons Attribution License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are cited.

Download Citation:  BibTeX | http://dx.doi.org/10.6180/jase.202609_32.037  

Download PDF

Traditional packaging image recognition methods rely on manual detection or manual feature extraction, which has the problems of low efficiency and poor adaptability, while existing deep learning models still face challenges when dealing with complex backgrounds, small target detection, and category imbalance. The purpose of this paper is to design an efficient and reliable product packaging image recognition system. By proposing a multi-scale attention fusion network (MSAFNet), this paper integrates a lightweight hybrid backbone network (combined with EfficientNet and ResNet), a dual attention module (DAM) to enhance feature focusing, an adaptive multi-scale feature pyramid (AFPN) to optimize multi-scale fusion, and a multi-task learning framework to jointly optimize detection and classification. The proposed model achieves an mAP@0.5 of 89.6% on the COCO dataset, outperforming the baseline by 4.4% while maintaining real-time performance at 40 FPS with only 8.9 M parameters. It demonstrates strong robustness with a low performance degradation rate of 10.8% and good generalization with only 6.1% cross-dataset degradation. These results highlight its effectiveness in balancing accuracy, efficiency, and scalability for industrial applications. The architecture design and system validation confirm its suitability for automated packaging detection tasks. However, performance under extreme occlusion remains a limitation and can be improved through future self-supervised learning approaches.

Keywords: convolutional neural network; products; packaging images; automatic identification; classification

  1. [1] S. Y. Chaganti, I. Nanda, K. R. Pandi, T. G. Prudhvith, and N. Kumar, (2020) “Image classification using SVM and CNN” Proceedings of the 2020 International Conference on Computer Science, Engineering and Applications (ICCSEA): 1–5. DOI: 10.1109/ICCSEA49143.2020.9132851.
  2. [2] Z. Liu, L. Sun, and Q. Zhang, (2022) “High similarity image recognition and classification algorithm based on convolutional neural network” Computational Intelligence and Neuroscience 2022: 2836486. DOI: 10.1155/2022/2836486.
  3. [3] K. V. Greeshma and J. V. Gripsy, (2020) “Image classification using HOG and LBP feature descriptors with SVM and CNN” International Journal of Engineering Research and Technology 8(4): 1–4. URL: https://d1wqtxts1xzle7.cloudfront.net/110347554.
  4. [4] M. Z. Alom, M. Hasan, C. Yakopcic, T. M. Taha, and V. K. Asari, (2020) “Improved inception-residual convolutional neural network for object recognition” Neural Computing and Applications 32(1): 279–293. DOI: 10.1007/s00521-018-3627-6.
  5. [5] T. Hassanzadeh, D. Essam, and R. Sarker, (2022) “EvoDCNN: An evolutionary deep convolutional neural network for image classification” Neurocomputing 488: 271–283. DOI: 10.1016/j.neucom.2022.02.003.
  6. [6] M. Tripathi, (2021) “Analysis of convolutional neural network-based image classification techniques” Journal of Innovative Image Processing 3(2): 100–117. DOI: 10.36548/jiip.2021.2.003.
  7. [7] M. A. Hasan, T. Bhargav, V. Sandeep, V. S. Reddy, and R. Ajay, (2024) “Image classification using convolutional neural networks” International Journal of Mechanical Engineering Research and Technology 16(2): 173–181. DOI: 10.1109/ICEEICT53079.2022.9768622.
  8. [8] M. Yang, P. Kumar, J. Bhola, and M. Shabaz, (2022) “Development of image recognition software based on artificial intelligence algorithm for the efficient sorting of apple fruit” International Journal of System Assurance Engineering and Management 13(Suppl. 1): 322–330. DOI: 10.1007/s13198-021-01415-1.
  9. [9] Z. Ren, F. Fang, N. Yan, and Y. Wu, (2022) “State of the art in defect detection based on machine vision” International Journal of Precision Engineering and Manufacturing–Green Technology 9(2): 661–691. DOI: 10.1007/s40684-021-00343-6.
  10. [10] Y. Liu, H. Pu, and D. W. Sun, (2021) “Efficient extraction of deep image features using convolutional neural network (CNN) for applications in detecting and analysing complex food matrices” Trends in Food Science and Technology 113: 193–204. DOI: 10.1016/j.tifs.2021.04.042.
  11. [11] Q. Zhang, Q. Yang, X. Zhang, Q. Bao, J. Su, and X. Liu, (2021) “Waste image classification based on transfer learning and convolutional neural network” Waste Management 135: 150–157. DOI: 10.1016/j.wasman.2021.08.038.
  12. [12] K. U. S. W. O. R. O. Adi, C. E. Widodo, A. P. Widodo, and U. S. Margiati, (2022) “Detection of foreign object debris (FOD) using convolutional neural network (CNN)” Journal of Theoretical and Applied Information Technology 100(1): 184–191. URL: https://www.jatit.org/volumes/Vol100No1/16Vol100No1.pdf.
  13. [13] S. Masood, U. Ahsan, F. Munawwar, D. R. Rizvi, and M. Ahmed, (2020) “Scene recognition from image using convolutional neural network” Procedia Computer Science 167: 1005–1012. DOI: 10.1016/j.procs.2020.03.400.
  14. [14] Y. Li, R. G. Zhou, R. Xu, J. Luo, and W. Hu, (2020) “A quantum deep convolutional neural network for image recognition” Quantum Science and Technology 5(4): 044003. DOI: 10.1088/2058-9565/ab9f93.
  15. [15] G. Chen, Q. Chen, S. Long, W. Zhu, Z. Yuan, and Y. Wu, (2023) “Quantum convolutional neural network for image classification” Pattern Analysis and Applications 26(2): 655–667. DOI: 10.1007/s10044-022-01113-z.
  16. [16] R. Basha, P. Pathak, M. Sudha, K. V. Soumya, and J. Arockia Venice, (2025) “Optimization of quantum dilated convolutional neural networks: Image recognition with quantum computing” Internet Technology Letters 8(3): e70027. DOI: 10.1002/itl2.70027.
  17. [17] H. C. V. Ngu and K. M. Lee, (2022) “Effective conversion of a convolutional neural network into a spiking neural network for image recognition tasks” Applied Sciences 12(11): 5749. DOI: 10.3390/app12115749.
  18. [18] V. Guizilini, R. Ambrus, S. Pillai, A. Raventos, and A. Gaidon, (2020) “3D packing for self-supervised monocular depth estimation” Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition: 2485–2494. URL: https://openaccess.thecvf.com/content_CVPR_2020/html/Guizilini_3D_Packing_for_Self-Supervised_Monocular_Depth_Estimation_CVPR_2020_paper.html.
  19. [19] I. J. Jacob and P. E. Darney, (2021) “Design of deep learning algorithm for IoT application by image-based recognition” Journal of ISMAC 3(3): 276–290. DOI: 10.36548/jismac.2021.3.008.
  20. [20] C. Zhang, M. Han, J. Jia, and C. Kim, (2024) “Packaging design image segmentation based on improved full convolutional networks” Applied Sciences 14(22): 10742. DOI: 10.3390/app142210742.
  21. [21] S. Aburass, A. Huneiti, and M. B. Al-Zoubi, (2022) “Classification of transformed and geometrically distorted images using convolutional neural network” Journal of Computer Science 18(8): 757–769. DOI: 10.3844/jcssp.2022.757.769.
  22. [22] D. Dai, Y. Li, Y. Wang, H. Bao, and G. Wang, (2022) “Rethinking the image feature biases exhibited by deep convolutional neural network models in image recognition” CAAI Transactions on Intelligence Technology 7(4): 721–731. DOI: 10.1049/cit2.12097.
  23. [23] M. Nauta, R. Van Bree, and C. Seifert, (2021) “Neural prototype trees for interpretable fine-grained image recognition” Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition: 14933–14943. URL: https://openaccess.thecvf.com/content/CVPR2021/html/Nauta_Neural_Prototype_Trees_for_Interpretable_Fine-Grained_Image_Recognition_CVPR_2021_paper.html.
  24. [24] Q. Cao, (2021) “The art of packaging: an investigation on modern packaging design and artistic thinking under the background of big data” Journal of Applied Science and Engineering 24(4): 807–812. DOI: 10.6180/jase.202110_24(5).0017.