Discriminative Regions and Adversarial Sensitivity in CNN-Based Malware Image Classification
Open Access
- 4 October 2025
- journal article
- research article
- Published by MDPI AG in Electronics
- Vol. 14 (19) , 3937
Abstract
The escalating prevalence of malware poses a significant threat to digital infrastructure, demanding robust yet efficient detection methods. In this study, we evaluate multiple Convolutional Neural Network (CNN) architectures, including basic CNN, LeNet, AlexNet, GoogLeNet, and DenseNet, on a dataset of 11,000 malware images spanning 452 families. Our experiments demonstrate that CNN models can achieve reliable classification performance across both multiclass and binary tasks. However, we also uncover a critical weakness in that even minimal image perturbations, such as pixel modification lower than 1% of the total image pixels, drastically degrade accuracy and reveal CNNs’ fragility in adversarial settings. A key contribution of this work is spatial analysis of malware images, revealing that discriminative features concentrate disproportionately in the bottom-left quadrant. This spatial bias likely reflects semantic structure, as malware payload information often resides near the end of binary files when rasterized. Notably, models trained in this region outperform those trained in other sections, underscoring the importance of spatial awareness in malware classification. Taken together, our results reveal that CNN-based malware classifiers are simultaneously effective and vulnerable to learning strong representations but sensitive to both subtle perturbations and positional bias. These findings highlight the need for future detection systems that integrate robustness to noise with resilience against spatial distortions to ensure reliability in real-world adversarial environments.Keywords
This publication has 28 references indexed in Scilit:
- Family Similarity-Enhanced Implicit Data Augmentation for Malware Classification2025
- A comparison of adversarial malware generatorsJournal of Computer Virology and Hacking Techniques, 2024
- On the Robustness of Image-Based Malware Detection Against Adversarial Attacks2024
- MIGAN: GAN for facilitating malware image synthesis with improved malware classification on novel datasetExpert Systems with Applications, 2023
- Generative adversarial networks and image-based malware classificationJournal of Computer Virology and Hacking Techniques, 2023
- Data augmentation and transfer learning to classify malware images in a deep learning contextJournal of Computer Virology and Hacking Techniques, 2021
- Image-Based malware classification using ensemble of CNN architectures (IMCEC)Computers & Security, 2020
- Using convolutional neural networks for classification of malware represented as imagesJournal of Computer Virology and Hacking Techniques, 2018
- Malware Visualization for Fine-Grained ClassificationIEEE Access, 2018
- Gradient-based learning applied to document recognitionProceedings of the IEEE, 1998