The world’s first 3D AI recognition system for large concrete structures
Last year Kiwa and Spotscale entered into a partnership agreement for the development of the inspection technology of the future, one that uses AI to predict the lifetime of concrete structures.
AI technology enables the detection of small cracks in concrete on large objects
AI (artificial intelligence) provides the opportunity to use high-resolution images taken by drones to map not only very minor but also major damage to inspected constructions (such as bridges, dams, power stations and water towers, for example). This technology can be used to monitor millions of individual flaws and defects as well as to categorise them in a way that is impossible to achieve with conventional inspection methods. The advantages for customers are dramatically enhanced quality, time savings and better control over the condition of concrete constructions.
Concrete is one of the most important building materials in the world.
In the early 1900s, reinforced concrete was considered a material with no or little need for maintenance and as good as everlasting. This has, however, since proven to be untrue and, as a result, Europe is now home to an incredible amount of defective infrastructure and damaged buildings.
Kiwa’s extensive experience and unique knowledge of consulting and the testing, inspection and analysis of, for example, concrete structures offers a huge amount of data for training the software’s AI.
Kiwa’s role is to produce quality assured training material encompassing relevant types of damage for the software’s AI as well as to validate inspection performance to ensure that it meets the established requirements for inspection and testing, which is something that Kiwa has solid experience of, such as from the nuclear power industry.
“Today, the automated image analysis and processing of enormous amounts of high-resolution imagery enable machine learning to find great numbers of small cracks in extremely large structures. However, this requires that this huge amount of information can be structured in a 3D model to provide an overview of the flaws and their sizes across the entire structure, which has been impossible until today,” says Ludvig Emgård, founder and CEO of Spotscale.
Thanks to Kiwa’s expert knowledge of and methods for concrete analysis, together with the automated machine learning methods and human input/adjustments from Spotscale, the results of the joint development work have the potential to change the very nature of how inspections and lifetime analysis of concrete infrastructure and buildings are conducted.
“Kiwa envisions great opportunities from our partnership with Spotscale. Their unique software and AI combined with Kiwa’s unique expertise and knowledge within concrete structures will provide real benefits to our customers and society in general. Working preventively and increasing efficiency reduces costs in both the short and the long term, says Erik Landgren, Country Manager of Kiwa Sweden AB”.
This partnership puts Kiwa and Spotscale to the fore of the world’s largest TIC companies when it comes to structural health monitoring of infrastructure.
About Kiwa
Today, Kiwa is one of the most successful companies operating in testing, inspection and certification and technical consulting. We are a global company with a local presence. We have strong customer relationships and deliver a high standard of service with a focus on safety, technology and sustainability. We have a broad offering encompassing services within process safety, building and lift safety, agriculture, the food industry and the construction industry. As a group, we operate in more than 40 countries and currently have about 10,000 employees throughout the world. In Sweden, we have around twenty offices, from Kiruna in the north to Malmö in the south, with some 900 employees working together for greater safety in society and agriculture. We contribute to increased safety within both industry and society.
About Spotscale
Spotscale is a technology company at the cutting edge of 3D, image analysis and machine learning, originating from decades of research within technical physics and computer science at, among other institutions, Linköping University. Thanks to the company’s unique ability to process large quantities of image material in the cloud, the condition of various types of building surfaces can be analysed in detail, providing valuable insights into wear and energy leaks. Spotscale was founded in 2013 and has 10 employees.
If you have any questions, please feel free to contact:
Milan Poznić, Head of Technology and Business Development, Kiwa, +46 104 793 194
milan.poznic@kiwa.com
For questions concerning the technology, please contact: Ludvig Emgård, Founder & CEO, Spotscale, +46 705 877 986
ludvig.emgard@spotscale.com