Embracing change in geo-data
Keith Owens, Commercial Director Americas Remote Sensing and Mapping, Fugro, talks to Danielle Kenneally
A SELF-CONFESSED geo-data enthusiast, Fugro’s Keith Owens is dedicated to the advancement in data management and cloud processes in order to enable better decision making. With the world’s landscape changing fast due to urban expansion and the increasing frequency of extreme weather events there is increasing demand for better geospatial data, the remote sensing and mapping and power commercial director wants to be the one people turn to for accurate geo-data that can help manage these changes effectively.
For him, geo-data users should be able to make better decisions and more effective preparations for what lies ahead. He believes with the advancement of data collection and data processes, we’ll see a rapid increase in everyday people and tools capturing and processing geo-data to populate our virtual worlds at scale – with ethical use of AI and human assistance making this accessible. Here he talks to the Geospatial Engineering’s Danielle Kenneally about the latest advancements, as well as the future advancements and what it is that makes him enjoy his role so much.
What is your role at Fugro?
My role is to design and implement Fugro’s commercial strategy for remote sensing and mapping, as well as the power industry in North and South America. In the past 12 months, we have been redeveloping our commercial strategy to respond efficiently to the fast-paced growing geo-data market. We’ve been active in building a commercial team that complements Fugro’s core values – determined to deliver, prepare for tomorrow, do what’s right, and build trust.
I’m very excited about this team, as we bring a unique characteristic to the commercial group, we all enjoy working with one-another, and we challenge each other to perform. I’m equally excited about our strategy to connect geo-data users with the technological innovations we’re developing.
How did you get to where you are today?
I’ve been in this business for over 27 years. At the start, I didn’t have the opportunity to learn the trade through a university. My introduction to aviation was fuelling airplanes at a local airport in the US. I then joined the mapping industry in 1996 as an aerial systems technician with an aviation group called EarthData which is now known as Fugro, out of Hagerstown, Maryland. We focused on aerial imagery for mapping services.
This was the same year our small team worked to develop one of the first wide-area airborne lidar sensors, known as an Aeroscan lidar system. My role at the time, was quite small. I occasionally helped with operating the sensor for testing purposes. I didn’t realise how big the airborne lidar market would be. My narrow view was that it would simply help with creating digital elevation models to generate ortho-imagery by removing the need for humans to manually compile the elevation values.
I’ve worked in just about every position in the business since then – I’ve operated aerial sensors for various projects throughout North and South America, set ground control for mapping projects, had a small stint of program management and frequency coordination for the development of a X- and P-band radar aircraft, compiled planimetrics, created and managed ortho-imagery projects, designed flight plans and cost estimates, did regional sales, business development and marketing, and now focused on our commercial team and strategy. It’s been a fun and diverse ride!
As a self-confessed ‘geo-data enthusiast’, what is it about your role that you enjoy and value most?
I enjoy and value our team, technology, and community. I’m really lucky to be working with a fantastic group of people. We are passionate about what we do. We are a diverse team with very different backgrounds, viewpoints, and experiences. That combination really helps us to remain relevant in the geo-data space and removes any opportunity for a stale work environment.
Since the beginning of my career, I’ve had the opportunity to be part of many technological advancements. It’s great to work for a company that embraces change and challenges the team to continue to find better ways to operate and create better products that will help solve our clients’ and also societal problems.
Our clients mean a lot to us. We have excellent relationships with geo-data users that share the same passion as we do. We know as the geo-data market continues to grow, it can be a crowded space at times. It means we also know there are many options when considering geo-data professionals to perform the work and so really value those that choose us.
How do you use this zeal to encourage others to get involved in the geospatial industry?
I had lunch with a professor from Shippensburg University in Pennsylvania, USA, not too long ago whose focus is on geospatial. During our chat, he mentioned our industry being a ‘found’ industry in that it wasn’t an industry most people aspire to be in when they are young or entering the university. That certainly made sense to me as I often find it difficult to explain what I do to my own family.
AI services like ChatGPT have several advantages over human interaction and operation. However, there are limitations to what AI can do, and it should not replace human interaction entirelySince most don’t really know much about our industry, I try and connect what we do, to what matters to them. I talk about projects that support positive change in the way humans operate. Projects that showcase how our geo-data helps for example, to improve stormwater management in flood-prone areas or how projects we work on support many sustainable operations and improve the impact on our natural environment.
When it comes to encouraging the younger generation to get involved in our industry, I like to share my experiences of travelling throughout North and South America doing survey work or flying in small aircraft, acquiring geo-data. By being out there, travelling, surveying the land, you have adventures that most people don’t have the opportunity to experience. Each new year is a new adventure in our business.
Working as a client advocate, you oversee the lifecycle of each project you work on to ensure its success, what project are you currently working on and how is it being used?
One of the more exciting innovations I’ve been working on is Sense.Lidar. Over the past handful of years, our geo-data team in North America, Europe and Australia have been combining expertise to improve the way we process lidar data. We recognised that our traditional methods for classifying clusters of lidar points was time-consuming, expensive, and typically littered with imperfections. By combining the brilliant minds in our Australia office with our experienced team in Europe and North America, we developed a machine learning process for classifying two points per square metre lidar data to a high degree of accuracy. We needed this to work on areas as small as 1km2 or as large as 300,000km2 It had to be scalable. This has been in operation for over a year now and we couldn’t be happier.
For example, last year we used Sense. Lidar to classify lidar points in the State of Texas covering 215,000 km2. The purpose was to take existing, federally sourced, United States Geological Survey (USGS) lidar data and add culverts, vegetation and building classifications to the dataset. This project converted the federal data to the Texas State specification. The data is used to support initiatives in Texas for dam safety, floodplain management and planning, feature extraction, water quality modelling, stream restoration potential analysis, vegetation analysis, forest management, building footprints, change detection, and emergency management services.
The State of Texas then also selected Fugro through a competitive bid scenario to, once again, classify the USGS data to the state specification. We’re currently working on roughly 60,000 square kilometres of lidar data. It’s exciting to know that, by using Sense.Lidar, we are expanding the use of the already-invested in lidar data.
The solution uses the AI technique of machine learning to produce results. How does this work and what are its benefits?
We value innovation, it’s exciting. We now use AI and cloud processes every day when managing large datasets.
I believe with the advancement of data collection and data processes, we’ll see a rapid increase in everyday people and tools capturing and processing geo-data to populate our virtual worlds at scale.However, what is equally important is our teams’ experience, understanding, handling, managing, and executing processes for getting the most out of the data. So, it’s not as simple as writing a clever algorithm, dropping data into a virtual bucket and out spits a perfectly crafted point cloud. We achieve the result by combining our teams experience with innovative processes.
To give a little insight into the processes, we carefully inspect the source data for quality and completeness, we fix systematic errors, we segment the project area by land cover type or category, we hand-craft training datasets to what a human would consider near-perfect classifications, we run the machine learning processes that we spent years developing, we inspect, fix, rerun, if necessary, quality control and deliver. The benefit to us, as a data acquisition company, is we improve the quality and accuracy of the product each time. The benefit to our client is that we improve project schedule and total cost.
AI services such as ChatGPT are creating ripples of change throughout the world and several industries – do you think the use of AI is more advantageous than human interaction and operation or are there and, additionally, should there be limitations to what it can do?
AI services like ChatGPT have several advantages over human interaction and operation. For instance, AI can process large volumes of data and information at a faster speed than humans. AI algorithms can also learn and improve their accuracy over time through machine learning, which is not possible for humans. Additionally, AI services are available 24/7 and can handle a large volume of requests without getting tired or fatigued, which can be challenging for human operators.
Since most don’t really know much about our industry, I try and connect what we do, to what matters to them. I talk about projects that support positive change in the way humans operateHowever, there are limitations to what AI can do, and it should not replace human interaction entirely. AI lacks the emotional intelligence and empathy that humans have, which can be crucial in some industries such as healthcare, customer service, and education. Additionally, AI systems can sometimes produce biased or incorrect results, which can have significant consequences.
It is essential to recognise the strengths and limitations of AI and use it appropriately in a complementary role alongside human interaction and operation. In some cases, AI can help humans perform their tasks more efficiently and effectively. Still, in other cases, human interaction is necessary to ensure empathy, understanding, and personalised attention to customers.
With this in mind, what could the future hold?
I know I’ve been in the business longer than some, less than others, but perhaps long enough to know we are just getting started. I’ve spent a career out travelling, acquiring data, processing data and delivering to clients. However, I believe with the advancement of data collection and data processes, we’ll see a rapid increase in everyday people and tools capturing and processing geo-data to populate our virtual worlds at scale – think vehicles and cell phones with lidar units attached. I believe AI will make this accessible.
However, it’s important for our geo-data community to operate ethically and responsibly to avoid potential problems generated out of making decisions from the AI-derived data. Human assistance is necessary.
What will be required to ensure this transformation is possible industrywide and for future generations to reap the benefits and go on to develop it further?
Investment, education, collaboration, regulation, and public engagement. The primary reason we’re able to offer Sense.Lidar to our clients is because Fugro invested in the idea. By investing in the idea, you can advance. Through education and training we can involve more people in AI technologies. Collaboration between private sector, academia and government, ensures the development of AI is done responsibly and safely. Regulation can prevent abuse and ensure its safe and ethical use. Public engagement is necessary to be sure AI technologies align with the communities’ values and expectations.
Keith Owens, Commercial Director Americas Remote Sensing and Mapping, Fugro, talks to Danielle Kenneally
Sense.Lidar is a registered trademark of Fugro