Guest Editorial: Deep Learning Assisted Visual IoT Technologies for Critical Infrastructure Protection

Abstract

The digitalization of business applications, including the critical infrastructure, promises transformative opportunities to society. IoT devices are one of the greatest technological advancements that have significantly changed the landscape of critical infrastructure applications. However, it presents an array of challenges in defining implementation standards and best practices to secure operational control systems associated with critical infrastructures. Further, critical infrastructures such as electric grids, railways, pipelines, and healthcare are susceptible to numerous threats. Effectively managing these infrastructures with conventional IoT devices is extremely time-consuming and expensive. Here, security becomes the primary concern as IoT devices are resource-constrained in nature, and they need cloud or edge computing infrastructures to process data and respond to infrastructure problems. In many cases, when dealing with long-distance surveillance of critical infrastructures, IoT devices face the problem of low channel capacity, which causes larger resource consumption in the delivery of images and videos. In this time gap, any significant deformation to the data sources may result in disastrous consequences. This creates the need for disruptive technologies to mitigate, adapt, and positively respond to various cyber threats across the critical infrastructure.