As AI and machine learning become more accessible, it opens up new opportunities for innovative applications in a wide variety of fields.
One example is geospatial technologies – with the help of machine learning, scientists working in fields like climate tracking, agriculture, forestry, disaster control, transportation, or logistics not only can save time but gather more accurate data and make better predictions.
In order to find out more about geospatial mapping technologies, Cybernews talked with Pierrick Poulenas, CEO & Co-Founder of Picterra – a software platform for training, deployment, and management of machine-learning models powering geospatial applications & business services.
Let’s go back to the very beginning of Picterra. What has the journey been like since your launch in 2016?
Since our founding in 2016, Picterra has come a long way. We started with the goal of making it easier for people to analyze satellite and drone imagery, and we've been working hard to achieve that ever since.
In the early days, we focused on developing our platform and refining our AI algorithms to enable our users to create their own advanced models. We participated in startup accelerators and incubators (including the ESA business incubation center) to help us grow our business and learn from others.
In 2018, we received our first pre-seed investment and launched our platform to the public, offering an easy-to-use self-service tool for image analysis. The platform quickly gained popularity and was used by a wide range of individuals, businesses, and organizations.
Over the years, we have continued to expand our capabilities, adding new features and functionality. We have also worked to make our platform more accessible, with a user-friendly interface and a SaaS pricing model. Further seed investment was secured in 2020, along with a new headquarters in Lausanne, Switzerland the following year. Just last year, in 2022, we secured $6.5 million in Series A funding.
Today, we are a leading provider of AI-powered Earth Observation image analysis tools, with customers in a wide range of industries, from agriculture to conservation to urban planning. We continue to innovate and develop new technologies with the goal of making it even easier for people to analyze and understand geospatial imagery. What makes Picterra unique is that we enable anyone, regardless of experience or skill set, to train their own deep learning models. We are proud of what we have accomplished so far, and we are excited to see what the future holds for Picterra.
Can you introduce us to your geospatial ML platform? What features make it stand out?
Picterra is a software platform for the training, deployment, and management of machine-learning models powering geospatial applications & business services. Picterra enables organizations to build scalable geospatial products and train their own deep-learning models with our geospatial MLOps platform.
The platform, which is entirely cloud-native, allows you to manage all your data in one place, create, train, and improve models in a collaborative environment, and bring them into production without needing additional resources from IT.
With built-in support for APIs and plugins, there are many options to build custom dashboards for clients or continue processing in other GIS software. Even entire SaaS platforms can be built by embedding Picterra’s powerful machine-learning capabilities.
The platform benefits our customers and partners by:
- Revolutionizing the way organizations approach the deployment and development of geospatial applications.
- Helping them unlock new opportunities and achieve their goals faster and more efficiently than ever before.
- Empowering them to grow while optimizing their operations and ensuring they impact the world in meaningful ways.
What are the main challenges associated with current geospatial mapping technologies?
With rocket prices having dropped fourfold in the past decade, we are currently in the midst of a “space renaissance”. It’s a period of massive technological innovation, driving significantly reduced costs and new capabilities. An explosion of new companies, along with a number of acquisitions and consolidations, has meant that the satellite industry is now producing 10 times more Earth imagery by area than five years ago.
This is unprecedented, but in order to produce accurate machine-learning models capable of quickly classifying crops or monitoring critical infrastructure, the quality of the imagery is key. Thus a key challenge for current geospatial mapping technologies is ensuring the future is all about image quality. This can be particularly challenging in areas with limited or outdated data. Though the costs of acquiring and processing geospatial data are coming down, they can still be high, particularly for high-resolution imagery and LiDAR data.
As geospatial mapping technologies become more advanced, there are concerns about how data is collected, used, and shared, particularly with regard to privacy and security. Finally, keeping geospatial mapping data up-to-date and accurate requires ongoing investment and maintenance, which can be challenging to sustain over time.
How did the recent global events affect your field of work?
Picterra stands in solidarity with Ukraine and its people during the recent conflict. We condemn the aggression and violation of Ukraine's territorial integrity and sovereignty. Satellite imagery has been used extensively to monitor and document the military activities of both sides, as well as to assess the damage caused by the conflict and assist in humanitarian missions.
At the start of the invasion in 2022, we made our geospatial platform and services available for free for any NGO or ally organization working on humanitarian missions related to the Ukraine invasion. We will help in any way we can by using our experience with satellite and aerial assessments for disaster relief and humanitarian aid.
In your opinion, why do certain companies hesitate to implement new and innovative solutions, despite all the technological advancements available nowadays?
Certain companies believe that implementing new and innovative solutions will require a significant investment and are afraid of the risks associated with change. Some may be unsure of how to integrate them into their existing systems or be hesitant to adopt new technologies due to concerns about regulatory compliance, data privacy, or other legal or ethical issues. Others may simply be unaware of the latest technological advancements and the potential benefits they can offer.
Whatever the reason, companies that are hesitant to adopt new and innovative solutions may struggle to keep pace with the rapidly changing business environment. At Picterra, we’re ready to help companies harness the power of geospatial data and advanced analysis to drive digital transformation and gain a competitive advantage. Our unique software platform enables anyone to train their own deep-learning models.
What other innovations do you hope to see become commonplace soon?
Geospatial machine learning is driving innovations across industries, with the greatest impact felt when it is able to answer urgently, enabling quick decisions to be made within crisis situations.
With the costs of disasters increasing, Nadine Alameh, CEO of the Open Geospatial Consortium (OGC), believes that “geospatial data is still not being used sufficiently to improve people’s lives”.
Geospatial technology such as what3words has already become synonymous with saving lives, and it’s our hope that innovations within geoAI have the same, if not greater, potential during 2023 and beyond.
For example, when earthquakes such as those in Turkey and Syria devastate communities, emergency management organizations are in a race against time to deploy responders and supplies. These organizations increasingly lean on data insights derived from satellite or aerial imagery to make the best plan for saving lives. As thousands of satellite images are taken after such events, data scientists can use geoAI to sort through the images and detect damaged buildings, houses, and other features.
In your opinion, which types of organizations would greatly benefit from geospatial applications?
We believe that In the next decade, every company in the world will be a geospatial-driven business. Geospatial applications have the potential to benefit a wide range of organizations across many industries, including helping:
Are there any precautions or security tools that you think need to be taken in this age of ever-changing technology?
Picterra is committed to ensuring the security and privacy of customer data and takes a proactive approach to security by implementing best practices and regularly reviewing and updating our security measures.
We have implemented several security measures to ensure that customer data is protected, including encrypting communications using SSL/TLS, offering a role-based access control system that controls who has access to different parts of the platform, and performing regular data backups to ensure that customer data is not lost due to accidental or intentional deletion or corruption.
Additionally, all customer data is stored in secure data centers from an ISO-27001-certified provider. We conduct regular security audits to identify and address any vulnerabilities or weaknesses in the platform.
Would you like to share what’s next for Picterra?
Within the geospatial industry, there is a growing demand for more sophisticated and automated AI-based solutions. We continue to develop and refine our leading and cutting-edge AI-powered platform to give our users access to faster, more accurate, and more efficient analysis of geospatial data.
Recently we announced powerful new data curation technology that allows users to get a better understanding of their datasets and improve model accuracy. This industry-first innovation gives GIS and data science teams an indication of where – based on image variability – they need to place new training areas in order to achieve maximum model accuracy. This is a powerful tool to speed up and streamline the annotation process and reduce the likelihood of false positives, and we are continuing to refine its functionality.
At Picterra, we are also focused on enabling solutions that address critical social and environmental issues such as climate change, disaster response, and conservation efforts. We want to help organizations navigate the complex challenges of climate change, promote sustainability, and secure our planet for future generations.