Opcode and Gray Scale Techniques for Classification of Malware Binaries

EasyChair Preprint 641

Abstract

Abstract--- In this study, we have used the image similarity technique to detect the unknown or new type of malware using CNN approach. CNN was investigated and tested with three types of datasets i.e. one from Vision Research Lab, which contains 9458 gray-scale images that have been extracted from the same number of malware samples that come from 25 different malware families, second was from Microsoft Malware Classification Challenge which contains 10868 Binary files that contains 9 different malware families and third was benign dataset which contained 3000 different kinds of benign software. Benign dataset and dataset from Microsoft Malware Classification Challenge were initially .EXE files which were converted into binary code and then converted into image files. We obtained a testing accuracy of 98% on Vision Research lab dataset and 97.6 accuracy on Microsoft Malware Classification Challenge dataset. 

Keyphrases: Convolutional Neural Network, deep learning, malware classification, malware detection

@booklet{EasyChair:641,
  author    = {Rajesh Kumar and Riaz Ullah Khan},
  title     = {Opcode and Gray Scale Techniques for Classification of Malware Binaries},
  howpublished = {EasyChair Preprint 641},
  year      = {EasyChair, 2018}}