Analysis of the Impact of Class Imbalance on the Quality of Three-Class Image Classification

Alexandra Zhdanova, Sergey Sablin, Alexandra Trukhova

Abstract


Recent advances in generative artificial intelligence have led to a rapid increase in synthetic imagery and deepfakes, posing new challenges to the reliability of visual information. Modern AI tools can produce highly realistic images of non-existent people, making it increasingly difficult to verify media content and maintain public trust in visual data. This study explores the task of classifying user avatars by their origin type (real, drawing, generated) under conditions of significant class imbalance. The research focuses on analyzing how uneven data distribution affects the accuracy of image classification. In this work, several deep learning architectures (including ResNet-50, MobileNetV3, EfficientNet-B0, and ConvNeXt-Tiny) were trained and evaluated using datasets from open sources. Experimental results demonstrate that when the dominant “drawing” class accounts for about 80% of the dataset, all models show a substantial decrease in macro-F1 scores during independent testing, indicating overfitting and reduced generalization. Among the evaluated models, ConvNeXt-Tiny exhibited the highest stability to class and domain shifts, while MobileNetV3 provided the most efficient balance between classification accuracy and computational cost. The findings highlight the importance of considering class imbalance when training and evaluating image classification models.

Full Text:

PDF (Russian)

References


Vedomosti. “Chislo sluchaev moshennichestva s ispol'zovaniem dipfeikov vyroslo v 2,3 raza” [Online]. Available: https://www.vedomosti.ru/technology/articles/2025/06/19/1118123-chislo-sluchaev-moshennichestva-s-ispolzovaniem-dipfeikov-viroslo Accessed: Oct. 30, 2025.

L. H. Singh, P. Charanarur, and N. K. Chaudhary, "Advancements in detecting Deepfakes: AI algorithms and future prospects – a review", Discover Internet of Things, vol. 5, p. 53, 2025. doi: 10.1007/s43926-025-00154-0.

T. Karras, S. Laine, and T. Aila, "A Style-Based Generator Architecture for Generative Adversarial Networks", in Proc. IEEE/CVF Conf. on Computer Vision and Pattern Recognition (CVPR), 2019, pp. 4396–4405. doi: 10.1109/CVPR.2019.00453.

R. Rombach, A. Blattmann, D. Lorenz, P. Esser, and B. Ommer, "High-Resolution Image Synthesis with Latent Diffusion Models", in Proc. IEEE/CVF Conf. on Computer Vision and Pattern Recognition (CVPR), 2022, pp. 10674–10685. doi: 10.1109/CVPR52688.2022.01042.

M. S. Rana, M. N. Nobi, B. Murali, and A. H. Sung, "Deepfake Detection: A Systematic Literature Review", IEEE Access, vol. 10, pp. 25494–25513, 2022. doi: 10.1109/ACCESS.2022.3154404.

A. Heidari, N. Jafari Navimipour, H. Dag, and M. Unal, "Deepfake detection using deep learning methods: A systematic and comprehensive review", WIREs Data Mining and Knowledge Discovery, vol. 14, no. 2, art. e1520, 2024. doi: 10.1002/widm.1520.

D. Namiot, E. Ilyushin, and I. Chizov, “Artificial Intelligence and Cybersecurity”, International Journal of Open Information Technologies, vol. 10, no. 9, pp. 135–147, 2022. [Online]. Available: https://www.injoit.org/index.php/j1/article/view/1402

D. Namiot, “Schemes of Attacks on Machine Learning Models”, International Journal of Open Information Technologies, vol. 11, no. 5, pp. 68–86, 2023. [Online]. Available: https://www.injoit.org/index.php/j1/article/view/1566

V. E. Ponomarev, “Forensic Examination of Digital Images Altered by AI: Applicability of New and Classical Analysis Methods”, Legal Science, no. 9, pp. 430–434, 2025. doi: 10.24412/2220-5500-2025-9-430-434.

S. P. Nikitenkova, “Application of Benford’s Law in Deepfake Image Detection”, Tomsk State University Journal of Control and Computer Science, no. 64, pp. 128–137, 2023. doi: 10.17223/19988605/64/13.

A. B. Mudrich and K. V. Ezhova, “Face Detection and Recognition in Video Surveillance Systems”, International Journal of Open Information Technologies, vol. 12, no. 3, pp. 82–87, 2024. [Online]. Available: https://www.injoit.org/index.php/j1/article/view/1798

N. Yu, L. Davis, and M. Fritz, "Attributing Fake Images to GANs: Learning and Analyzing GAN Fingerprints", in Proc. IEEE/CVF Int. Conf. on Computer Vision (ICCV), 2019, pp. 7555–7565. doi: 10.1109/ICCV.2019.00765.

S.-Y. Wang, O. Wang, R. Zhang, A. Owens, and A. A. Efros, "CNN-Generated Images Are Surprisingly Easy to Spot... for Now", in Proc. IEEE/CVF Conf. on Computer Vision and Pattern Recognition (CVPR), 2020, pp. 8692–8701. doi: 10.1109/CVPR42600.2020.00872.

R. Corvi, D. Cozzolino, G. Zingarini, G. Poggi, K. Nagano, and L. Verdoliva, "On the Detection of Synthetic Images Generated by Diffusion Models", in Proc. IEEE Int. Conf. on Acoustics, Speech and Signal Processing (ICASSP), 2023, pp. 1–5. doi: 10.1109/ICASSP49357.2023.10095167.

Z. Wang et al., "DIRE for Diffusion-Generated Image Detection", in Proc. IEEE/CVF Int. Conf. on Computer Vision (ICCV), 2023, pp. 22388–22398. doi: 10.1109/ICCV51070.2023.02051.

U. Ojha, Y. Li, and Y. J. Lee, "Towards Universal Fake Image Detectors that Generalize Across Generative Models", in Proc. IEEE/CVF Conf. on Computer Vision and Pattern Recognition (CVPR), 2023, pp. 24480–24489. doi: 10.1109/CVPR52729.2023.02345.

M. Zhu et al., "GenImage: A Million-Scale Benchmark for Detecting AI-Generated Image", in Advances in Neural Information Processing Systems 36 (NeurIPS 2023), Datasets and Benchmarks Track, 2023. [Online]. Available: https://proceedings.neurips.cc/paper_files/paper/2023/hash/f4d4a021f9051a6c18183b059117e8b5-Abstract-Datasets_and_Benchmarks.html

J. J. Bird and A. Lotfi, "CIFAKE: Image Classification and Explainable Identification of AI-Generated Synthetic Images", IEEE Access, vol. 12, pp. 15642–15650, 2024. doi: 10.1109/ACCESS.2024.3356122.

J. M. Johnson and T. M. Khoshgoftaar, "Survey on deep learning with class imbalance", Journal of Big Data, vol. 6, art. 27, 2019. doi: 10.1186/s40537-019-0192-5.

S. Rezvani and X. Wang, "A broad review on class imbalance learning techniques", Applied Soft Computing, vol. 143, art. 110415, 2023. doi: 10.1016/j.asoc.2023.110415.

Y. Zhang, B. Kang, B. Hooi, S. Yan, and J. Feng, "Deep Long-Tailed Learning: A Survey", IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 45, no. 9, pp. 10795–10816, 2023. doi: 10.1109/TPAMI.2023.3268118.

T.-Y. Lin, P. Goyal, R. Girshick, K. He, and P. Dollár, "Focal Loss for Dense Object Detection", IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 42, no. 2, pp. 318–327, 2020. doi: 10.1109/TPAMI.2018.2858826.

Y. Song, T. Wang, P. Cai, S. K. Mondal, and J. P. Sahoo, "A Comprehensive Survey of Few-shot Learning: Evolution, Applications, Challenges, and Opportunities", ACM Computing Surveys, vol. 55, no. 13s, art. 271, 2023. doi: 10.1145/3582688.

K. Zhou, Z. Liu, Y. Qiao, T. Xiang, and C. C. Loy, "Domain Generalization: A Survey", IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 45, no. 4, pp. 4396–4415, 2023. doi: 10.1109/TPAMI.2022.3195549.

K. He, X. Zhang, S. Ren, and J. Sun, "Deep Residual Learning for Image Recognition", in Proc. IEEE Conf. on Computer Vision and Pattern Recognition (CVPR), 2016, pp. 770–778. doi: 10.1109/CVPR.2016.90.

A. Howard et al., "Searching for MobileNetV3", in Proc. IEEE/CVF Int. Conf. on Computer Vision (ICCV), 2019, pp. 1314–1324. doi: 10.1109/ICCV.2019.00140.

M. Tan and Q. V. Le, "EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks", in Proc. 36th Int. Conf. on Machine Learning (ICML), PMLR, vol. 97, 2019, pp. 6105–6114. [Online]. Available: https://proceedings.mlr.press/v97/tan19a.html

Z. Liu, H. Mao, C.-Y. Wu, C. Feichtenhofer, T. Darrell, and S. Xie, "A ConvNet for the 2020s", in Proc. IEEE/CVF Conf. on Computer Vision and Pattern Recognition (CVPR), 2022, pp. 11966–11976. doi: 10.1109/CVPR52688.2022.01167.

Kaggle. [Online]. Available: https://www.kaggle.com/ (accessed: Aug. 12, 2026).

I. Loshchilov and F. Hutter, "Decoupled Weight Decay Regularization", in Proc. 7th Int. Conf. on Learning Representations (ICLR), 2019. [Online]. Available: https://openreview.net/forum?id=Bkg6RiCqY7

R. R. Selvaraju, M. Cogswell, A. Das, R. Vedantam, D. Parikh, and D. Batra, "Grad-CAM: Visual Explanations from Deep Networks via Gradient-Based Localization", International Journal of Computer Vision, vol. 128, no. 2, pp. 336–359, 2020. doi: 10.1007/s11263-019-01228-7.

L. van der Maaten and G. Hinton, "Visualizing Data using t-SNE", Journal of Machine Learning Research, vol. 9, pp. 2579–2605, 2008. [Online]. Available: https://jmlr.org/papers/v9/vandermaaten08a.html


Refbacks

  • There are currently no refbacks.


Abava  Кибербезопасность ИТ-КОНГРЕСС ВМК МГУ 2026 СНЭ

ISSN: 2307-8162