Enhanced Cnn Is Used For Mal Image Anomaly Detection And Classification

Abstract

Malware is a major cause of concern for today's businesses. In order to attack their targets, attackers and hackers are continually developing new malicious software. Security firms are putting in their best efforts to protect against malware attacks, but they are unable to do so due to the millions of new malware samples detected each month. As a result, novel approaches like deep learning are needed. Classic computer programs face many challenges and difficulties in identifying objects for a variety of reasons, including lighting, viewpoint, deformation, and segmentation. Convolutional Neural Networks (CNNs) are a deep learning approach to tackling the image classification problem, or what we refer to as computer vision problems, because classic computer programs face many challenges and difficulties in identifying objects for a variety of reasons, including lighting, viewpoint, deformation, and segmentation. Convolutional Neural Networks (CNNs) are a deep learning technique for solving picture categorization issues, also known as computer vision problems. CNN are organized in three-dimensional structures with width, height, and depth as differentiating characteristics. In the case of images, the height represents the image's height, the width represents the image's width, and the depth represents the number of RGB channels. For this new model, the work utilizes CNNs to create a malware classifier