Malware Binary Image Classification Using Convolutional Neural Networks


  • John Kiger Rowan University, Glassboro, NJ, USA
  • Shen-Shyang Ho Rowan University, Glassboro, NJ, USA
  • Varhid Heydari Rowan University, Glassboro, NJ, USA



malware analysis, malware classification, malware visualization, convolutional neural networks, deep learning


The persistent shortage of cybersecurity professionals combined with enterprise networks tasked with processing more data than ever before has led many cybersecurity experts to consider automating some of the most common and time-consuming security tasks using machine learning. One of these cybersecurity tasks where machine learning may prove advantageous is malware analysis and classification. To evade traditional detection techniques, malware developers are creating more complex malware. This is achieved through more advanced methods of code obfuscation and conducting more sophisticated attacks. This can make the manual process of analyzing malware an infinitely more complex task. Furthermore, the proliferation of malicious files and new malware signatures increases year by year. As of March 2020, the total number of new malware detections worldwide amounted to 677.66 million programs. In 2020, there was a 35.4% increase in new malware variants over the previous year. This paper examines the viability of classifying malware binaries represented as fixed-size grayscale using convolutional neural networks. Several Convolutional Neural Network (CNN) architectures are evaluated on multiple performance metrics to analyze their effectiveness at solving this classification problem.