EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Addressing Data Scarcity in Malware Classification via Pixel-Level Synthetic Image Generation

Mounika Krishna Teja Karumudi, Fabio Di Troia

Electronics · 30.06.2026 · 15 (13) · 2848

DOI: 10.3390/electronics15132848 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Electronics, online erschienen am 30.06.2026. Verfasst von Mounika Krishna Teja Karumudi und 1 weiteren Autor:innen. Die Arbeit nennt 28 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Deep learning-based malware classification using image representations has emerged as a highly effective paradigm for threat detection. However, training robust neural networks is frequently bottlenecked by data scarcity and severe class imbalances in real-world repositories. This study investigates the viability of using an autoregressive PixelCNN framework to synthesize high-fidelity, class-specific malware images to augment limited training distributions. Utilizing the benchmark Malimg dataset, we systematically evaluate a Convolutional Neural Network (CNN) classifier across varying ratios of synthetic-to-authentic data under strict data scarcity constraints (ranging from 10 to 80 authentic samples per family). Our experimental results reveal that while PixelCNN successfully replicates intricate, byte-level micro-textures, classifiers trained exclusively on synthetic data experience catastrophic performance degradation, yielding an accuracy of just 3%. Crucially, however, the introduction of a minimal authentic data anchor (15% to 20%) restores functional decision boundaries, immediately elevating classification accuracy up to 72%. Furthermore, performance saturates rapidly once the training matrix reaches a 50/50 synthetic-to-authentic split, achieving up to 82% classification accuracy, rendering it highly competitive with the 89% accuracy upper bound of a fully authentic baseline. These findings demonstrate an exceptional degree of data efficiency, proving that generative autoregressive augmentation can halve the authentic data collection burden in cybersecurity workflows provided a minor, real-world baseline anchor is preserved.

Zitieren

APA

Karumudi, M. K. T., & Troia, F. D. (2026). Addressing Data Scarcity in Malware Classification via Pixel-Level Synthetic Image Generation. Electronics, 15(13), 2848. https://doi.org/10.3390/electronics15132848

Vancouver

Karumudi MKT, Troia FD. Addressing Data Scarcity in Malware Classification via Pixel-Level Synthetic Image Generation. Electronics. 2026;15(13):2848. doi:10.3390/electronics15132848

BibTeX
@article{Karumudi2026,
  author = {Karumudi, Mounika Krishna Teja and Troia, Fabio Di},
  title = {Addressing Data Scarcity in Malware Classification via Pixel-Level Synthetic Image Generation},
  journal = {Electronics},
  year = {2026},
  volume = {15},
  number = {13},
  pages = {2848},
  doi = {10.3390/electronics15132848}
}
Kostenlos starten →