With its TWINSPECT software solution, Kassel-based Twinsity aims to make inspections digital, safe and efficient. Infrastructures and buildings worldwide are in an increasingly critical state and the need for intelligent solutions to replace manual structural inspections is becoming more pressing. Drone technology has already overcome some of the difficulties of conventional methods, but the data from a drone inspection could not be used to its full potential. This is where TWINSPECT comes in, by linking the high-resolution 3D model of the object with the original images on an intelligent cloud platform to digitally perform a comprehensive inspection and analyze the data with technological support. All stakeholders in the project can securely manage and use the data collaboratively in one place and thus gain in-depth insights into the object.
For an efficient inspection, various tools are already available to mark, classify and measure damaged areas. The Distr@l grant will now enable the development and integration of an AI function into the existing software to automatically detect and analyze damage to objects (such as cracks in an asphalt surface, vegetation damage or graffiti on buildings). For this purpose, the already existing Twinsity system uses artificial neural networks, which are obtained from existing data sets on the individual damage classes. On this basis, in the long term, it will also be possible to compare models over time and the associated prediction of damage developments. The vision is to extend the TWINSPECT software to create a holistic tool in the field of “asset lifecycle management”.
For more information on the funding project, visit: https://lidia-hessen.de/projekte/21_0017_4b/