Malicious Image Detection Using Convolutional Neural Network

Abstract

In previous era, malware attacks have achieved serious heights. As information technology field strengthens, the activities of cyber-criminals are also updated. Cybercriminals always look for those methods to attack which are not much suspicious. Attackers started to use approaches like steganography to conceal the scripts. With the wide use of images on social media and other platforms like World Wide Web (WWW), attackers started to embed the malwares in images. With the growth of malware attacks through images, it is high time to introduce a technique which would detect the malicious images. Proposed study aims the detection of images which are concealed with different scripts. We used a dataset of JPEGs, containing 1100 malicious and 1100 benign images to employ the detection method. Our method of malicious image detection would help everyone to prevent the malware attacks which are carried through images.