Simply enter your keyword and we will help you find what you need.
How we enriched 3D point cloud data sharing with AI
XomniaCasesHow we enriched 3D point cloud data sharing with AI
"We have known Xomnia for years now, and from the onset we have only had good experiences with the experts from Xomnia. Steven Woudenberg, Manager Advanced Analytics, Asset.Insight.
More about the case:
What they needed
Asset.Insight. is the data service provider for VolkerWessels, a Dutch construction company aiming to create sustainable, innovative and future-oriented living environments. Like most construction companies, VolkerWessels is increasingly relying on 3D point cloud data for accurate and fast records of the 3D geometries within their projects.
However, this data is often being stored on hard drives and used for a single goal. By migrating to cloud infrastructure, VolkerWessels would make data sharing easier amongst their 120 local operating companies worldwide and increase the searchability and security of their data.
The stored data could also help VolkerWessels companies to advance their asset management. For example, they could predict potential equipment failures. Since they manage assets related to rail, roads, and water, this would increase both the availability of the infrastructure and the safety for company employees and for the general public.
Asset.Insight. created a proof of concept for the project and turned to Xomnia to help turn it into a scalable solution. The key to the success of this project was enabling the vast amount of data to be easily and conveniently discoverable by VolkerWessels employees once it was migrated to the cloud. This saves time and reduces the expenses by preventing duplicate work.
What we did
In collaboration with the Asset.Insight. team, we created a platform for users within and outside of VolkerWessels to safely store large 3D datasets. Through the webportal or API’s, users can search through these data sets and can download (parts) of the files.
The next step is to enable the user to not only download 3D point clouds, but also apply state-of-the-art algorithms on the point clouds on the platform. For example, if a client would like to know the location and height of all trees in an area, the user can select that area and the algorithm is applied on the available 3D point cloud data in that area.
How it works
With Azure Active Directory integration, clients can seamlessly use the same credentials they use within their company to authenticate to Asset.Insight.’s services. Azure Data Lake Storage is used to provide a scalable and fault tolerant storage solution with isolation between each clients’ provisioned storage.
The various processing tasks to be carried out are scheduled on a Kubernetes cluster for scalability and efficient use of resources. We utilised Cesium and their 3D tiles standard in order to display available data in 3D, overlayed on existing maps and terrain models.
What we achieved
The immediate business impact is two-fold. Sharing the 3D point clouds through a platform, which is always accessible, enables the clients to easily find and re-use the 3D point cloud data for all projects. The solution does not limit itself to 3D point clouds; any type of files can be uploaded and be made available. Furthermore, clients can have easy access to the results of algorithms which can be applied on the uploaded data.
At the moment, Asset.Insight. is attracting more users to the platform to test it and store their data. Once the product is fully developed, it will be possible to utilise the data to accurately predict asset failures. The state of the art algorithms used for this can then be integrated into the platform. This has enormous potential for further future-proofing Asset.Insight.