Sponsor: National Science Foundation (2019-2021)
Despite the strategic role that computer-aided design (CAD) files play in AM, designers develop, store, and use the CAD models as they do in any other manufacturing method. This study will research a novel method of imparting new functionalities in CAD models by converting them to the frequency domain using lossless algorithms. The frequency domain representation allows searching the files for specific design features of interest, which is not feasible in the current CAD file formats. Transformation to the frequency domain opens up possibilities for developing new compression and encryption methods. Analysis will be conducted to determine the fundamental principles that lead to lossless conversion of CAD files to the frequency domain and develop compression and encryption methods for the files. Wavelet transform and dynamic time warping will be among the methods used for implementing search functionalities. Dynamic window selection and scaling methods will be used to perform feature search with and without scaling of the object size. These algorithms are based on an exact match procedure and do not rely on tuning or learning the algorithm. The project will help industries that are using AM methods, for example, the aerospace, automotive, medical and military equipment industries. Development of these unique capabilities will by provide new capabilities for design, search, and security. The project will involve post-doctoral fellows and graduate students in the research.
Participants:
- Nikhil Gupta (PI)
- Ramesh Karri (Co-PI)
- Hammond Pearcce (Post-doc)
- Wenjin Li (M.S. Candidate)
- Gary Mac (Ph.D. Candidate)
Publications:
- Wenjin Li, Gary Mac, Nektarios Georgios Tsoutsos, Nikhil Gupta, Ramesh Karri, Computer aided design (CAD) model search and retrieval using frequency domain file conversion, Additive Manufacturing, 2020, 101554
- Mouris, D., Gouert, C., Gupta, N. and Tsoutsos, N. G. Peak your Frequency: Advanced Search of 3D CAD Files in the Fourier Domain IEEE Access, 141481-141496.