@conference{dhm 31779, author = {Yordan Kyosev, Tino Kühn, Ann-Malin Schmidt}, title = {Automatic generation of partially homogenized FEM human body models based on 3D body scan data}, volume = {7}, year = {2022}, url = {https://pubs.lib.uiowa.edu/dhm/article/id/31779/}, issue = {1}, doi = {10.17077/dhm.31779}, abstract = {<p>Human safety and sitting comfort in the car depend on a complex set of interactions between the human body, its clothing, and the automotive environment in the meaning of sit elements, belt, solid parts, and air. The design of any functional clothing requires elastic human body models, too, but the detailed FEM models for car crash simulations are unnecessarily accurate, computationally intensive, and not practicable for the clothing development process. For correct simulation of the mechanical interaction, full-scale FEM models of humans with enough suitable accuracy and complexity are required.</p><p>This work presents the development steps and current state of an algorithm for automatic solid FEM mesh generator for human bodies, based on 3D scan data. The data from the 3D scanner is used for two purposes: (a) for detection of the main sizes of the bones of the human body and (b) the surface mesh is used as a basis for the building of the inner solid layers of the skin, inner soft structures, and the bones.</p><p>The automatically created solid mesh can then be used for the evaluation of the mechanical interaction between the human body, its clothing, and the environment within commercial or open-source FEM software. The model does not accurately build the interaction within the human body (between bones, muscles, and skin), but allows evaluation of the mechanical interactions and the sitting comfort for different body types and sizes in significantly reduced time.</p>}, month = {8}, pages = {2 pages}, keywords = {human body,FEM model,3D body scanning,partial homogenization,clothing-body interaction}, publisher={University of Iowa}, journal = {Proceedings of the 7th International Digital Human Modeling Symposium} }