Using AI to “See” the World More Clearly

August 15, 2018

Visual intelligence is one of the most underutilized technologies in rail right now. Visual Intelligence (or VI) is using AI to process video information in order to allow systems to “see.” “Seeing” is the ability to detect and interpret information from visible light to build a representation of the surrounding environment. Video footage is captured all day, every day, but nothing is done with this data to improve our systems. Railroads have invested in this technology over the past decade, but the data collected is only used when there is an accident, and then the rest is thrown away.

At Wi-Tronix, we’re focusing on what we can do with VI to detect problems and prevent accidents. VI has the potential to proactively mitigate risk and increase safe autonomy on the road, which will become increasingly important as we see more and more autonomous vehicles on the road (check out my latest video on AVs here). This technology is already being widely used in other industries, like social media, and we in the transportation industry need to reframe the way we think about VI to improve our rail systems.

Part of the reason we couldn’t fully utilize the vast amount of data collected by video before was the lack of a processing system. This is where AI comes into play. AI allows us to process hundreds of thousands of hours of video data that we couldn’t before. AI is made for repeatable, structured tasks and gives employees the freedom to do what they are great at— unpredictable, unstructured tasks that require a human touch.

AI is essential to fully analyze the video captured and leverage the data collected, but many people are wary of AI. I get it, people don’t always trust technology and are worried about job losses. To those skeptics, I ask: is it truly efficient to have people analyze video? You would need tens of thousands of employees staring at screens all day to process all the data. People aren’t doing these jobs right now for a reason, and it would be unreasonable to try and have them do what technology can do better. AI is designed for this sort of task and is very efficient at it. AI increases the breadth of possibilities for the rail industry, and with the right team, AI can be implemented safely and efficiently.

VI has the capacity to bring attention to problems that may otherwise go unnoticed, from logistical issues to potential dangers. This technology is real, and it exists today. With all the benefits of VI, you may be saying: “This is a no-brainer, so why aren’t people using this technology already?? The answer is because many people don’t know or don’t believe that this is possible. (But it is.)

We can’t wait for others to implement new technology for us; it’s up to us to leverage the technology to move the rail industry forward. The wheel has been invented for us. What we need to do is put it on the right axle to make it work for rail.

Originally published on LinkedIn