Analytics in the IoT public sector: Perfecting with computer vision
Optimization
Interestingly, the correlation between the greater efficiency, decreased costs, and enhanced mission effectiveness of such timely management of IoT data is exemplified in the private sector. The granular nature of the integration of IoT data provided by virtualization technologies is ideal for devising long-term strategies on these real-time inputs, thereby expanding the IoT’s utility from low-latency action to decision making based on historical data as well. Shah referenced an emergent use case in which logistics ships are transporting international cargo. “As these ships become more autonomous, you don’t necessarily have a whole crew manning that ship.” In this instance, computer vision is helpful for detecting anomalies on the seas related to navigation, security, unexpected weather, and more.
Its combination with other IoT data is essential “when you’re taking, let’s say, a cargo from China to the U.S. and you’re navigating inclement weather and those types of things,” Shah ventured. There are IoT sensors on the regulators and the pumps ascertaining how much fuel is being used and the route being taken. Once the cargo is transported, analytics on all of this data in conjunction with that of other historical data—including previous trips, different cargo amounts, or the same launching point to different destinations—is invaluable for optimizing each aspect of additional journeys. “It can help them monitor costs associated with routes, fuel, logistics, and all those types of things,” Shah acknowledged. This method also enables these companies to reach their objectives faster and more effectively than they could without such analytics. The same approach has horizontal applications for any number of use cases in the private sector.
Beyond IoT
The natural synthesis of IoT and computer vision technologies is almost certain to serve both of these dimensions of the data landscape, as well as benefit the public and private sectors. When coupled with video data, the continuous streaming of IoT devices practically warrants computer vision in almost any scenario in which, as Bennett stated, there simply aren’t enough employees to “stare at the cameras” in perpetuity. Simultaneously, the use of IoT presents some of the more profound opportunities to increase organizational effectiveness and efficiency, as the foregoing use cases in authentication, anomaly detection, proactive maintenance, and optimization indicate. As Shankar commented, the ability to funnel the results of such distributed analytics in a centralized manner is crucial to both “the data integration and data delivery side” for prompt action.
Regardless, the most notable facet of the overall utility of computer vision may be that it is by no means restricted to the realm of IoT. Its benefits in diminishing costs, maximizing employee productivity, and taming the decentralized data landscape are just as applicable to internal use cases such as medical examinations or auto insurance damage estimates. In these applications and others, it assists the primary mission of the private sector by not only minimizing costs, but by also increasing revenue. As Bennett noted, the former stems from amplifying efficiency, the latter from inflating organizational effectiveness.