Refine
Year of publication
- 1999 (232) (remove)
Institute
- Fachbereich Wirtschaftswissenschaften (50)
- Fachbereich Medizintechnik und Technomathematik (38)
- Fachbereich Elektrotechnik und Informationstechnik (31)
- Fachbereich Energietechnik (28)
- Fachbereich Maschinenbau und Mechatronik (24)
- Fachbereich Luft- und Raumfahrttechnik (17)
- Fachbereich Chemie und Biotechnologie (16)
- Fachbereich Bauingenieurwesen (12)
- Solar-Institut Jülich (8)
- Fachbereich Gestaltung (6)
Has Fulltext
- no (232) (remove)
Document Type
- Article (153)
- Book (47)
- Conference Proceeding (15)
- Patent (7)
- Part of a Book (6)
- Doctoral Thesis (1)
- Habilitation (1)
- Report (1)
- Review (1)
Keywords
- 20 fossil-fueled power plants (1)
- ceramics (1)
- filtration (1)
- gas flow (1)
- hot gas cleanup (1)
Die Anfertigung wissenschaftlicher Arbeiten und ihre Präsentation - ein komprimierter Einstieg
(1999)
"The invention relates to a pot spinning device (1) comprising a thread guide (9) which can be moved backwards and forwards and introduced into an opening (8) on the end side of the pot (7) so as to deposit a thread on the inside wall of a spinning pot. In order to improve upon pot spinning devices of this kind, the spinning pot is connected to an electric drive mechanism (10), mounted in magnetic bearings and provided with a gastight housing. Said housing encompasses the spinning pot (7), is open in the area where there is an opening in the end side of the pot and is provided with a seal (19) to seal off the intermediate area (17) between the housing and spinning pot. The housing is also connected to a supply of light-gas via a feed line (18), thereby enabling a light-gas atmosphere to be maintained in the intermediate area."
Aim of the AXON2 project (Adaptive Expert System for Object Recogniton using Neuml Networks) is the development of an object recognition system (ORS) capable of recognizing isolated 3d objects from arbitrary views. Commonly, classification is based on a single feature extracted from the original image. Here we present an architecture adapted from the Mixtures of Eaqerts algorithm which uses multiple neuml networks to integmte different features. During tmining each neural network specializes in a subset of objects or object views appropriate to the properties of the corresponding feature space. In recognition mode the system dynamically chooses the most relevant features and combines them with maximum eficiency. The remaining less relevant features arz not computed and do therefore not decelerate the-recognition process. Thus, the algorithm is well suited for ml-time applications.