SavedAdd your journal

Journals / Black Sea Journal of Engineering and Science / 2026 / Cilt: 9 Sayı: 3

A Hybrid Deep Learning Approach with Spatial Attention Mechanism for Visual-Based Malware Detection in IoT Security

Kubilay Ayturan(GAZİ ÜNİVERSİTESİ)0000-0001-9406-4694
Pages
1256–1268
Publication Date
2026-05-15
🏛️Bibliographic metadata and full text access for this article are provided via DergiPark.
DergiPark Record ↗
✨

AI Summary & Key Takeaways

AI Insight
⚡ TL;DR(Too Long; Didn't Read)

The findings reveal that visual analysis methods enhanced with attention mechanisms demonstrate higher performance than standard convolutional neural networks and offer an effective solution for detecting next-generation cyber threats.

📌 3 Key Takeaways

🎯
Objective & Hypothesis:The rapid proliferation of embedded technologies and the Internet of Things (IoT) has triggered a dramatic escalation in the frequency and complexity of malicious activities targeting such networks.
📊
Key Findings & Data:Conventional detection techniques relying on signatures often fall short when countering obfuscation and packing strategies commonly employed by cybercriminals.
💡
Conclusion & Impact:The findings reveal that visual analysis methods enhanced with attention mechanisms demonstrate higher performance than standard convolutional neural networks and offer an effective solution for detecting next-generation cyber threats.
🔬Design:Araştırma Makalesi
📍Scope:9,339 samples
⚙️Variables:Malware detection · Deep learning · Image processing

Abstract

The rapid proliferation of embedded technologies and the Internet of Things (IoT) has triggered a dramatic escalation in the frequency and complexity of malicious activities targeting such networks. Conventional detection techniques relying on signatures often fall short when countering obfuscation and packing strategies commonly employed by cybercriminals. I present a novel transfer learning-driven neural network architecture which examines malicious software through the conversion of raw binary codes to generate grayscale visual representations. The original value of the study lies in the Spatial Attention mechanism integrated into the standard ResNet-50 architecture, which enables the network to prioritize salient texture patterns of malicious code. To validate the technique and its generalization capability, comprehensive experiments were conducted utilizing both the grayscale Malimg benchmark archive (9,339 samples, 25 families) and the RGB-based MaleVis dataset (26 families). Empirical findings demonstrate that the proposed hybrid framework yielded exceptional classification accuracies of 99.36% on Malimg and 97.25% on MaleVis, proving its robust dataset-independence. The findings reveal that visual analysis methods enhanced with attention mechanisms demonstrate higher performance than standard convolutional neural networks and offer an effective solution for detecting next-generation cyber threats.

Özet

The rapid proliferation of embedded technologies and the Internet of Things (IoT) has triggered a dramatic escalation in the frequency and complexity of malicious activities targeting such networks. Conventional detection techniques relying on signatures often fall short when countering obfuscation and packing strategies commonly employed by cybercriminals. I present a novel transfer learning-driven neural network architecture which examines malicious software through the conversion of raw binary codes to generate grayscale visual representations. The original value of the study lies in the Spatial Attention mechanism integrated into the standard ResNet-50 architecture, which enables the network to prioritize salient texture patterns of malicious code. To validate the technique and its generalization capability, comprehensive experiments were conducted utilizing both the grayscale Malimg benchmark archive (9,339 samples, 25 families) and the RGB-based MaleVis dataset (26 families). Empirical findings demonstrate that the proposed hybrid framework yielded exceptional classification accuracies of 99.36% on Malimg and 97.25% on MaleVis, proving its robust dataset-independence. The findings reveal that visual analysis methods enhanced with attention mechanisms demonstrate higher performance than standard convolutional neural networks and offer an effective solution for detecting next-generation cyber threats.

🔖

Cite This Article

Ayturan, K. (2026). A Hybrid Deep Learning Approach with Spatial Attention Mechanism for Visual-Based Malware Detection in IoT Security. Black Sea Journal of Engineering and Science, Cilt: 9 Sayı: 3, 1256–1268. https://doi.org/10.34248/bsengineering.1851726

📄 Full Text PDF Viewer

📄

PDF görüntüleyici hazırlanıyor...

References (27)

  1. Akhtar, M. S., & Feng, T. (2022). Detection of malware by deep learning as CNN-LSTM machine learning techniques in real time. Symmetry, 14(11), 2308. https://doi.org/10.3390/sym14112308
  2. Almobaideen, W., Abu Alghanam, O., Abdullah, M., Hussain, S. B., & Alam, U. (2025). Comprehensive review on machine learning and deep learning techniques for malware detection in android and IoT devices. International Journal of Information Security, 24(3), 110. https://doi.org/10.1007/s10207-025-01027-x
  3. Alohali, M. A., Alahmari, S., Aljebreen, M., Asiri, M. M., Miled, A. B., Albouq, S. S., Alrusaini, O., & Alqazzaz, A. (2025). Two stage malware detection model in internet of vehicles (IoV) using deep learning-based explainable artificial intelligence with optimization algorithms. Scientific Reports, 15(1), 20615. https://doi.org/10.1038/s41598-025-00269-y
  4. Aslan, O., & Yilmaz, A. A. (2021). A new malware classification framework based on deep learning algorithms. IEEE Access, 9, 87936–87951. https://doi.org/10.1109/ACCESS.2021.3089586
  5. Benbrahim, H., & Behloul, A. (2024). Malware classification on Malimg using MobileNet and LSTM for efficient detection. In 2024 1st International Conference on Innovative and Intelligent Information Technologies (IC3IT) (pp. 1–6). IEEE. https://doi.org/10.1109/IC3IT63743.2024.10869415
  6. Copiaco, A., El Neel, L., Nazzal, T., Mukhtar, H., & Obaid, W. (2023). A neural network approach to a grayscale image-based multi-file type malware detection system. Applied Sciences, 13(23), 12888. https://doi.org/10.3390/app132312888
  7. Damaševičius, R., Venčkauskas, A., Toldinas, J., & Grigaliūnas, Š. (2021). Ensemble-based classification using neural networks and machine learning models for Windows PE malware detection. Electronics, 10(4), 485. https://doi.org/10.3390/electronics10040485
  8. El-Sayed, R., El-Ghamry, A., Gaber, T., & Hassanien, A. E. (2021). Zero-day malware classification using deep features with support vector machines. In 2021 Tenth International Conference on Intelligent Computing and Information Systems (ICICIS) (pp. 311–317). IEEE. https://doi.org/10.1109/ICICIS52592.2021.9694256
  9. Gibert, D., Mateu, C., & Planes, J. (2020). The rise of machine learning for detection and classification of malware: Research developments, trends and challenges. Journal of Network and Computer Applications, 153, 102526. https://doi.org/10.1016/j.jnca.2019.102526
  10. Jia, L., Yang, Y., Tang, B., & Jiang, Z. (2023). ERMDS: A obfuscation dataset for evaluating robustness of learning-based malware detection system. BenchCouncil Transactions on Benchmarks, Standards and Evaluations, 3(1), 100106. https://doi.org/10.1016/j.tbench.2023.100106
  11. Johny, J. A., Asmitha, K. A., Vinod, P., Radhamani, G., Rehiman, K. A. R., & Conti, M. (2025). Deep learning fusion for effective malware detection: Leveraging visual features. Cluster Computing, 28(2), 135. https://doi.org/10.1007/s10586-024-04723-w
  12. Kaggle repository. (2025). Malimg original dataset. https://www.kaggle.com/datasets/ikrambenabd/malimg-original (Erişim tarihi: 28 Aralık 2025)
  13. Kalash, M., Rochan, M., Mohammed, N., Bruce, N. D. B., Wang, Y., & Iqbal, F. (2018). Malware classification with deep convolutional neural networks. In 2018 9th IFIP International Conference on New Technologies, Mobility and Security (NTMS) (pp. 1–5). IEEE. https://doi.org/10.1109/NTMS.2018.8328749
  14. Kılıç, U. (2023). Deep learning-based airspeed estimation system for a commercial aircraft [Ticari uçaklar için derin öğrenme tabanlı hava hızı tahmin sistemi]. Journal of Aeronautics and Space Technologies, 16(2), 20–35. https://jast.msu.edu.tr/index.php/JAST/article/view/539
  15. Kilic, U., Villareal-Valderrama, F., Ayar, M., Ekici, S., Amezquita-Brooks, L., & Karakoc, T. H. (2024). Deep learning-based forecasting modeling of micro gas turbine performance projection: An experimental approach. Engineering Applications of Artificial Intelligence, 130, 107769. https://doi.org/10.1016/j.engappai.2023.107769
  16. Kim, H., & Kim, M. (2024). Malware detection and classification system based on CNN-BiLSTM. Electronics, 13(13), 2539. https://doi.org/10.3390/electronics13132539
  17. Liaqat, A., Shahid, U., Shah, I., Kaleem, A., & Riaz, A. (2025). Deep learning-based malware detection using independent stream analysis of RGB and grayscale images. In 2025 International Conference on Artificial Intelligence, Computer, Data Sciences and Applications (ACDSA) (pp. 1–5). IEEE. https://doi.org/10.1109/ACDSA65407.2025.11166223
  18. Naeem, H., Ullah, F., Naeem, M. R., Khalid, S., Vasan, D., Jabbar, S., & Saeed, S. (2020). Malware detection in industrial internet of things based on hybrid image visualization and deep learning model. Ad Hoc Networks, 105, 102154. https://doi.org/10.1016/j.adhoc.2020.102154
  19. Nataraj, L., Karthikeyan, S., Jacob, G., & Manjunath, B. S. (2011). Malware images: Visualization and automatic classification. In Proceedings of the 8th International Symposium on Visualization for Cyber Security (pp. 1–7). ACM. https://doi.org/10.1145/2016904.2016908
  20. Nisa, M., Shah, J. H., Kanwal, S., Raza, M., Khan, M. A., Damaševičius, R., & Blažauskas, T. (2020). Hybrid malware classification method using segmentation-based fractal texture analysis and deep convolution neural network features. Applied Sciences, 10(14), 4966. https://doi.org/10.3390/app10144966
  21. Sihwail, R., Omar, K., & Zainol Ariffin, K. A. (2018). A survey on malware analysis techniques: Static, dynamic, hybrid and memory analysis. International Journal on Advanced Science, Engineering and Information Technology, 8(4-2), 1662–1671. https://doi.org/10.18517/ijaseit.8.4-2.6827
  22. Su, X., Zhang, D., Li, W., & Zhao, K. (2016). A deep learning approach to android malware feature learning and detection. In 2016 IEEE Trustcom/BigDataSE/ISPA (pp. 244–251). IEEE. https://doi.org/10.1109/TrustCom.2016.0070
  23. Sudhakar, & Kumar, S. (2021). MCFT-CNN: Malware classification with fine-tune convolution neural networks using traditional and transfer learning in Internet of Things. Future Generation Computer Systems, 125, 334–351. https://doi.org/10.1016/j.future.2021.06.029
  24. Menon, V., Kumaravelu, V. B., Kumar C, V., Rammohan, A., Sunilchinnadurai, Venkatesan, R., Hai, H., & Poongundranselvaprabhu. (2025). AI-powered IoT: A survey on integrating artificial intelligence with IoT for enhanced security, efficiency, and smart applications. IEEE Access, 13, 50296–50339. https://doi.org/10.1109/ACCESS.2025.3551750
  25. Vasan, D., Alazab, M., Wassan, S., Naeem, H., Safaei, B., & Zheng, Q. (2020). IMCFN: Image-based malware classification using fine-tuned convolutional neural network architecture. Computer Networks, 171, 107138. https://doi.org/10.1016/j.comnet.2020.107138
  26. Vu, D., Nguyen, T., Nguyen, T. V., Nguyen, T. N., Massacci, F., & Phung, P. H. (2020). HIT4Mal: Hybrid image transformation for malware classification. Transactions on Emerging Telecommunications Technologies, 31(11), Article e3789. https://doi.org/10.1002/ett.3789
  27. Zhang, X., Wu, K., Chen, Z., & Zhang, C. (2021). MalCaps: A capsule network based model for the malware classification. Processes, 9(6), 929. https://doi.org/10.3390/pr9060929