An Efficient Convolutional Neural Network for Fashion Image Classification on the Fashion-MNIST Dataset
Article Sidebar
Main Article Content
Abstract
The e-commerce fashion industry requires automatic product classification systems to efficiently manage thousands of items. This study aims to develop a Convolutional Neural Network (CNN) model to classify three fashion product categories: Bag, Dress, and Sneaker on the Fashion MNIST dataset. The dataset used consists of 900 images with a split of 720 training images and 180 testing images. The CNN architecture was developed with three convolutional blocks equipped with Batch Normalization and Dropout, along with data augmentation to improve generalization on limited datasets. The model was trained using Adam optimizer with a learning rate of 0.0001 and early stopping strategy. Evaluation results show the model achieved a testing accuracy of 97.22% with the highest validation accuracy of 100%. Per-class performance shows Sneaker achieved perfect accuracy of 100%, Bag achieved 98.33%, and Dress achieved 93.33%, with only 5 errors out of 180 predictions. The developed model can help the e-commerce industry automate fashion product classification processes, improve operational efficiency, and reduce product categorization errors.
Downloads
Article Details
Section
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
How to Cite
References
[1]. H. Xiao, K. Rasul, and R. Vollgraf, "Fashion-MNIST: A Novel Image Dataset for Benchmarking Machine Learning Algorithms," arXiv preprint arXiv:1708.07747, 2017. Available: https://arxiv.org/abs/1708.07747
[2]. Y. LeCun, Y. Bengio, and G. Hinton, "Deep learning," Nature, vol. 521, no. 7553, pp. 436-444, 2015. DOI: 10.1038/nature14539
[3]. "Convolutional Neural Networks and Vision Transformers for Fashion MNIST Classification: A Literature Review," arXiv preprint arXiv:2406.03478, 2024. Available: https://arxiv.org/abs/2406.03478
[4]. "Evaluation of CNN Models with Fashion MNIST Data," Iowa State University Digital Repository, 2019. Available: http://dr.lib.iastate.edu/handle/20.500.12876/95513
[5]. "CNNs Architectures: LeNet, AlexNet, VGG, GoogLeNet, ResNet and more," Medium Analytics Vidhya. Available: https://medium.com/analytics-vidhya/cnns-architectures-lenet-alexnet-vgg-googlenet-resnet-and-more-666091488df5
[6]. K. Simonyan and A. Zisserman, "Very Deep Convolutional Networks for Large-Scale Image Recognition," International Conference on Learning Representations (ICLR), 2015. Available: https://arxiv.org/abs/1409.1556
[7]. S. Ioffe and C. Szegedy, "Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift," Proceedings of the 32nd International Conference on Machine Learning, vol. 37, pp. 448-456, 2015.
[8]. K. He, X. Zhang, S. Ren, and J. Sun, "Deep Residual Learning for Image Recognition," IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 770-778, 2016.
[9]. H. Jin, Q. Song, and X. Hu, "Auto-Keras: An Efficient Neural Architecture Search System," Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, 2019. Available: https://arxiv.org/abs/1806.10282
[10]. A. Dosovitskiy et al., "An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale," International Conference on Learning Representations (ICLR), 2021.
[11]. "Enhancing Fashion Classification with Vision Transformers," Electronics, vol. 12, no. 20, 2023. Available: https://www.mdpi.com/2079-9292/12/20/4263
[12]. A. Krizhevsky, I. Sutskever, and G. E. Hinton, "ImageNet Classification with Deep Convolutional Neural Networks," Advances in Neural Information Processing Systems, vol. 25, pp. 1097-1105, 2012.
[13]. "Transfer Learning using Pre-Trained AlexNet Model and Fashion-MNIST," Towards Data Science. Available: https://towardsdatascience.com/transfer-learning-using-pre-trained-alexnet-model-and-fashion-mnist-43898c2966fb/
[14]. "Transfer Learning on Fashion MNIST Using PyTorch," GeeksforGeeks. Available: https://www.geeksforgeeks.org/transfer-learning-on-fashion-mnist-using-pytorch/
[15]. N. Srivastava, G. Hinton, A. Krizhevsky, I. Sutskever, and R. Salakhutdinov, "Dropout: A Simple Way to Prevent Neural Networks from Overfitting," Journal of Machine Learning Research, vol. 15, pp. 1929-1958, 2014.
[16]. "Privacy-Enhanced Federated Learning with Differential Privacy on Fashion MNIST," Nature Scientific Reports, 2025. Available: https://www.nature.com/articles/s41598-025-12575-6
[17]. D. P. Kingma and J. Ba, "Adam: A Method for Stochastic Optimization," International Conference on Learning Representations (ICLR), 2015. Available: https://arxiv.org/abs/1412.6980
[18]. C. Shorten and T. M. Khoshgoftaar, "A survey on Image Data Augmentation for Deep Learning," Journal of Big Data, vol. 6, no. 60, 2019. DOI: 10.1186/s40537-019-0197-0
[19]. M. Bhobé, "Classifying Fashion with a Keras CNN (achieving 94% accuracy) - Part 1," Medium, 2019. Available: https://medium.com/@mjbhobe/classifying-fashion-with-a-keras-cnn-achieving-94-accuracy-part-1-1ffcb7e5f61a
[20]. C. Masch, "Evaluation of Fashion-MNIST with a Simple CNN," GitHub Repository, 2018. Available: https://github.com/cmasch/zalando-fashion-mnist
[21]. A. S. Rana, "Fashion-MNIST CNN Benchmarking Accuracy upto ~98%," GitHub Repository, 2019. Available: https://github.com/AkhilSinghRana/Fashion-MNIST-CNN
[22]. P. Ahirwar and M. Soni, "CNN Model for Image Classification on MNIST and Fashion-MNIST Dataset," Journal of Scientific Research, vol. 64, no. 2, pp. 374-384, 2020. Available: https://www.bhu.ac.in/research_pub/jsr/Volumes/JSR_64_02_2020/51.pdf
[23]. "Fashion-MNIST using Deep Learning with TensorFlow Keras," CloudxLab Blog, 2019. Available: https://cloudxlab.com/blog/fashion-mnist-using-deep-learning-with-tensorflow-keras/
[24]. J. Deng, W. Dong, R. Socher, L. Li, K. Li, and L. Fei-Fei, "ImageNet: A Large-Scale Hierarchical Image Database," IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 248-255, 2009.
[25]. K. Balabanova, D. Dudin, and D. Matveev, "State-of-the-Art Results with the Fashion-MNIST Dataset," Mathematics, vol. 12, no. 20, 2024. Available: https://www.mdpi.com/2227-7390/12/20/3174
[26]. I. Goodfellow, Y. Bengio, and A. Courville, "Deep Learning," MIT Press, 2016. Available: http://www.deeplearningbook.org