Trauth, DanielDanielTrauthStanke, JoachimJoachimStankeFeuerhack, AndreasAndreasFeuerhackBergs, ThomasThomasBergsMattfeld, PatrickPatrickMattfeldKlocke, FritzFritzKlocke2022-03-052022-03-052018https://publica.fraunhofer.de/handle/publica/25625810.1016/j.promfg.2018.07.2802-s2.0-85063781265In fine blanking the sheared edge's quality is of major importance. As the sheared edge needs to transmit process forces and precisely align parts, attributes like die rolls, tears and tear-offs need to be eliminated. Currently, these attributes are manually determined at the end of the process chain, which makes a complete correlation between influencing factors and the attributes hardly possible. If it would be possible to create a real-time evaluation of the attributes, not only unknown correlations would be found, also an instant process adaption could be made, optimizing the quality of the sheared edge. Therefore, this contribution focusses on the development of an edge computing approach.enfine blankingsheared edges qualityimage processingedge computingdeep learning658670journal article