Pixstart’s Chief Science Officer Dr Lydwine Gross outlines how the company combines satellite data with economic and climate information to offer geostrategy and geomarketing services at near real-time at low cost.
With more than 15 years of experience in space sector major industrial companies, the three partners of the Pixstart company use satellite data to bring together science and digital, which are key areas of expertise in geo-intelligence services.
The start-up from Toulouse, France, is positioning itself in the development of new added value services with satellite data. They use spatial data (remote sensing radiometry, geolocation, mapping, etc.), open data (Insee, GIS data, etc.), and Artificial Intelligence, through partnerships with industry in the space sector.
Speaking to SciTech Europa Quarterly, Dr Lydwine Gross – Pixstart’s Chief Science Officer and one of the company’s partners – outlines how the company combines satellite data with economic and climate information to offer geostrategy and geomarketing services at near real-time at low cost.
As a European company, what were the biggest challenges in starting and scaling up? What are your thoughts on European-level support for start-ups?
Starting up is really the most difficult part. Indeed, in Europe there is quite a lot of support for companies which are looking to scale up (for us, this is available from national governments and organisations such as the European Space Agency (ESA), and CNES (Centre national des Etudes Spatial) which are currently our main funders), and yet while the first year for any company is going to be the most challenging, with business centres and investors wanting to see solid figures for the first financial year, there is a distinct lack of support here, and we would really like to see this change.
At Pixstart, we were able to use personal finances to ensure our success throughout this initial period, and this has enabled us to grow and to go on to attract funding from external sources.
Have you found that what is termed the ‘Big Science market’, which includes organisations such as ESA, is difficult to enter and/or navigate for small companies such as Pixstart?
No, this has not been a particular problem for us. At Pixstart, we have been working on big satellite data for over 20 years; it is the lifeblood of the company. As such, this experience and level of expertise means that we are able to demonstrate a level of competence and have been able to quickly develop a suite of services which have valuable real-world applications, and the Big Science market has been very receptive to that.
How does your own experience working on climate change satellite data feed into what you are now tasked with at Pixstart?
Both I and my colleagues have been monitoring how the Earth is changing for the last 20 years. My own previous experience means that I am able to competently use numerous different kinds of satellites which can retrieve a lot of different kinds of information about the state of the Earth system.
These include changes in:
- The atmosphere;
- The water; and
- On the surface
The satellites we use measure all of these different states, meaning that I am able to make an inventory of everything that is present on Earth and, moreover, I am able to monitor the energy that is passing from one system to another. For example, when looking at a city, I can explore how the gases being generated by that urban space are expanding in the atmosphere; or I can check the water resources or the humidity of the soil and so see if and how the vegetation is being affected.
The continuous improvement in satellite technologies means that we are able to do more that we have ever been able to do before. In the past, we were only able to check these variables and parameters concerning continents, oceans, and forests, and so on, at the regional scale. Now, however, we can obtain a resolution of just 10 metres.
This means that we can monitor a city; we can monitor human activities; and we can precisely monitor things like agriculture. This enhanced resolution is enabling us to provide new opportunities to private companies so that they can better comprehend the world in which they are working and interacting.
How is Pixstart working to assist the private sector?
One of our first services (Poolspot) involves monitoring the location of residential swimming pools. We combine this information with economic and climate data in our geomarketing exercises so as to create strategic maps which not only detail where existing pools are located, but also where future swimming pools will be built. These maps are of use to all those working in the swimming pool sector, including those responsible for their sale, construction, and maintenance. This allows them to do targeted marketing campaigns to swimming pool owners or future owners and hopefully grow their business.
A second service involves the monitoring of how construction sites are developing over time (this is called ‘Buildspot’ (see Fig. 1), thanks to which we are finalists of the Copernicus masters, see: www.copernicus-masters.com/hallof-fame/#finalists-2018). This is important as once a construction site has been identified and future work has been declared, it can often take an undetermined period of time for the work to actually begin. We therefore provide information to those working in the construction sector so that they are aware of when building work begins and are therefore able to more effectively market their products and services.
These kinds of services are very important in a world where the marketplace is increasingly global and which continues to change at an increasing rate. In such an environment, it is difficult – and will become increasingly so – for companies to survive if they cannot look to work outside of their own country.
Pixstart offers these businesses a global vision of their market. Regarding the swimming pool market, we can provide a pool builder based, for instance, in Europe, with data about swimming pools in another location, such as Brazil, meaning that they have the capacity to expand the scope of their operations. Indeed, this service would also enable them to considerably reduce the cost of such an attempted expansion, not least because they no longer need to physically travel to another country in order to gauge the market.
Essentially, what we offer is a global geostrategy and geomarketing services at near real-time at low cost.
How important is AI to the services Pixstart offers?
I have been working on the application of AI to satellite data for more than 20 years (my PhD was actually on this subject), since this concept was exported to Europe from the USA, where it was first developed.
At Pixstart, we work to make operational the algorithms we have been working on for some 15 years because we feel that the time is now right to take this step forwards. Of course, using AI with neural networks is not a new concept, but we feel that with the evolution of the computer sciences and the fact that we are now able to calibrate a neural network and apply it to satellite data, coupled with the capacity to adequately handle Big Data, means that now is the perfect time to commercialise the research we have been involved in for almost two decades into applications which benefit society.
Are there any plans to extend the range of services in the future?
We provide an end-to-end service; our clients only see the results they have requested, which we have gathered from our satellite data – map of the pools, the state of the construction sites, and other new services (for instance Woodwatch, the calculation of wood production, see Fig. 2). In this sense, we are a data intelligence company.
In the future, we will go on to use satellite data as an open data source combined with other sources of data – economic-, societal-, climate-related, and, for instance, client data. However, while our first clients are private companies, moving forwards we want to engage with national governments as well.
An example of this is the work we are hoping to do with African nations thanks to our partnership contract with Geomatica, whereby we will use our expertise and our ability to work at the continent scale to develop a product to help governments to identify ephemeral water sources in order to help agriculture.
Currently, Pixstart helps managers in the private sector to make informed decisions, and we want to expand this further.