Trojan Malware Image Pattern Classification

The malicious data’s are grouped into various types of viruses such as Trojan, Trojan downloader and so on. The cyber security issues are increasing day to day. It is a challenging task for the network security and antivirus designers. Trojan malware family has been used to detect the newly arrived malicious data to a known group. It is a well popular research problem to detect and classify the Trojan viruses. This research issue is solved using image processing techniques. To identify the texture patterns of malware images is tedious task, because of the similarities present in the various malware families. In this proposed algorithm Gabor wavelet is used for key of feature extraction method. The dataset Malimng consists of the 25 malware variant families for each class max 300–1000 samples are there. The experimental results are analyzed compared with two classifications such as KNN and SVM. The texture patter classification accuracy is improved and false positive rate is decreased. The KNN gives accuracy 89.11% and SVM gives 75.11%.