Research on Malicious Code Homology Analysis Method Based on Texture Fingerprint Clustering
Zhao Xiaolin, Zhang Yiman, Xuhui Li, Chen Quanbao · 2018
In recent years, the number of malware and malicious code variants has increased dramatically, which results in serious problems in the network security field. However, the existing method is not accurate and efficient. In order to improve the method, we study and find that there are commonalities between the malicious software and image recognition. Therefore, we introduce the image feature extraction technology and propose a malicious code homology analysis method based on the texture fingerprint clustering. The binary malicious programs which are no source code can be visualized by this way, and the image texture fingerprint information be analyzed so that the type and family of malicious programs can be found more efficiently and accurately. In addition, this paper designs and implements a prototype experiment to verify the validity of the method. The results show that the malicious code homology analysis method based on the texture fingerprint clustering is effective and accurate to analyze the homology.