Refurbishing Utilities’ Security Leveraging AI and Video Analytics



How can Utilities’ Leaders ensure security using AI and Video Analytics?

There is a revolution brewing in the technology industry. Deep learning, an artificial intelligence (AI) method that uses deep or multi-layer neural networks is dramatically enhancing computer vision and video analytics. So that deep learning-based computer vision can beat human capability in rapidly identifying objects in images.

With available video technology, deep learning can help us process a considerable amount of data, which has been too great for humans alone to process. IBM is taking this technology to the next step.

IBM has announced the incorporation of PowerAI Vision and IBM Intelligent Video Analytics (IVA), enabling organizations to build customized AI models for recognizing objects or activities of interest that can be easily integrated into IVA for easy model deployment.

Training artificial intelligence models using deep learning is an arduous task, which requires preparing and labeling the data, selecting the appropriate neural network model, and picking and turning hyper-parameters during training to enhance AI model accuracy.

The PowerAI Vision software suite uses automatic machine learning methods to automatically train deep learning models for object detection and image classification. PowerAI Vision takes a data set of pictures or video frames labeled with the objects to identify, selects an appropriate neural network model, and trains the model by choosing the right hyper-parameters for best AI model accuracy. This software can be used to ensure utilities’ security.


Key Capabilities of Video Analytics Ensuring Utilities’ Security

Advanced Detection: Accurate alarming based on specific targets types and actions

Situational Awareness: The ability to convey the crucial details of a security event fast to understand map-based format.

Autonomous Sensor Control: Automated steering of cameras to an event location and subsequent hands free video tracking of a suspect.

Real-time Target Location: Real-time location information of incidents and real-time location tracking of potential intruders.

Gartner’s latest survey on CIO shows that 94% of Energy, Oil, and Gas Utilities’ IT leaders prioritize location and cybersecurity, 38% are investing in artificial intelligence and machine learning, and 17% in the internet of things (IoT), including sensor-based video technologies.

Energy, Oil & Gas utilities face increasing pressure to minimize costs through automation while dealing with revenue shortfalls due to regulatory compliance costs and suspension of disconnections due to customers’ hardships during the COVID-19 pandemic.

IT departments of Energy, Oil & Gas are leading the convergence of cyber and physical security, relying on video cameras. Its sensors collect thermal, infrared, and night vision video streaming, which provides raw data feed to video analytics systems to find anomalies in operations or security.

In 2020, through many challenges, utilities are learning how crucial it is to have a 360-degree view of their physical and cybersecurity operations. As per Black & Veatch’s Smart Utilities Report, only 54% of North America’s CIOs have a security operations center.

Preventing the perimeters of their remote and headquarter locations is the top priority of 94% of North American CIOs surveyed by Gartner. Physical and cyber-attack on any remote location are the biggest concern, and probably forcing a utility shut down could be a solution to recover.

That is why real-time monitoring of remote locations using several cameras to capture thermal, heat, and night-vision data and analyzed it in real-time using video analytics is essential. Leaders in this sector consider PowerAI Vision and Twenty20 Solutions that encompass remote video security, remote access control, portable surveillance, temporary deployments, automation and control, and real-time remote monitoring. Their use of video analytics, artificial intelligence, and machine learning can recognize potential assets, machinery, and remote site security risks in real-time.

Utilities’ CIOs, CISO, and IT leaders need to consider integrating AI and video analytics into their enterprise security strategy. Every threat surface of a Utility business can be monitored 24×7 using video cameras as sensors when combined with software, secured to every endpoint.