An Optimized and Hybrid Framework for Image Processing Based Network Intrusion Detection System

The network infrastructure has evolved rapidly due to the ever-increasing volume of users and data. massive number online devices forced transform facilitate operational necessities consumers. Among these necessities, security is prime significance. Network intrusion detection systems (NIDS) are among most suitable approaches detect anomalies assaults on a network. However, keeping up with requirements quite challenging constant mutation in attack patterns by intruders. This paper presents an effective prevalent framework for NIDS merging image processing convolution neural networks (CNN). proposed first converts non-image data from traffic into images then further enhances those using Gabor filter. classified CNN classifier. To assess efficacy recommended method, four benchmark datasets i.e., CSE-CIC-IDS2018, CIC-IDS-2017, ISCX-IDS 2012, NSL-KDD were used. approach showed higher precision contrast recent work mentioned datasets. Further, method compared well-known methods NIDS.