Deep Learning and Binary Representational Image Approach for Malware Detection
Varsha Hemant Patil, Shinit Shetty, Anushka Tawte, Sanskruti Wathare · 2023
This research proposes the use of a “binary representational image” for malware classification. The proposed solution used a deep learning approach to classify malware. There are several methods for analyzing and studying malware. The technique based on image processing and image classification is a novel approach to analyzing malware. Malware can be categorized according to its family types, making it easier to assess the damage the file could have or has already caused. A total of 35 categories of image families are used for classification. Convolution Neural Networks (CNNs) are used in the proposed image-based classification system to detect malicious content. With CNN, the proposed method achieved an accuracy of 92.73