Final Report Summary - DETECT (New product design and engineering technologies based on next-generation computed tomography)
The principal objectives of the DETECT project were to develop new basic hardware components and algorithms for fast and accurate Three-dimensional computed tomography (3D-CT) scanning for parts of any complexity and large dimensions, new algorithms for automated CT image processing, fast surface and solid modelling, as well as methodologies for the integration of the new technologies into the rapid product development and manufacturing process. A next-generation CT prototype was set up for validation and testing. The CT data acquisition had to be at least 10 times faster compared to current scanners with line detector systems, providing the same level of accuracy. Particular attention had also to be paid to make these new technologies well-known to the European key industries.
In parallel to the CT development, there was good progress in the development of new software tools for data processing. Algorithms were elaborated and tested to process CT data in the most automatic way in order to use the generated data for direct product engineering. These include software tools for the automatic determination of adapted gray value thresholds to segment the CT voxel data or tools to optimise the resulting triangle meshes for simulation applications without reconstructing a CAD model.
Moreover, notable results were achieved on processing large point clouds. Existing CT to Computer-aided design (CAD) conversion solutions were not adopted to the huge amount of data which was generated by CT-scans (millions of points). Consequently, new software tools were developed in order to be able to load, visualise and, of course, process these very large data sets. For the next steps in the conversion of CT data to high-quality CAD models, new automatic feature detection algorithms based on curvature analysis and best fit algorithms were applied, optimised curve networks were generated and finally the surface model was built automatically. Tests by the end-users in the project proved the capabilities of the different software tools and their integration in the rapid product development process chain. In addition e-learning and demonstration tools for different kind of target groups have been built in order to spread the industrial CT technology and its application fields all over Europe.
In parallel to the CT development, there was good progress in the development of new software tools for data processing. Algorithms were elaborated and tested to process CT data in the most automatic way in order to use the generated data for direct product engineering. These include software tools for the automatic determination of adapted gray value thresholds to segment the CT voxel data or tools to optimise the resulting triangle meshes for simulation applications without reconstructing a CAD model.
Moreover, notable results were achieved on processing large point clouds. Existing CT to Computer-aided design (CAD) conversion solutions were not adopted to the huge amount of data which was generated by CT-scans (millions of points). Consequently, new software tools were developed in order to be able to load, visualise and, of course, process these very large data sets. For the next steps in the conversion of CT data to high-quality CAD models, new automatic feature detection algorithms based on curvature analysis and best fit algorithms were applied, optimised curve networks were generated and finally the surface model was built automatically. Tests by the end-users in the project proved the capabilities of the different software tools and their integration in the rapid product development process chain. In addition e-learning and demonstration tools for different kind of target groups have been built in order to spread the industrial CT technology and its application fields all over Europe.
