Conference Proceeding
Refine
Year of publication
Institute
- Fachbereich Elektrotechnik und Informationstechnik (310)
- Fachbereich Energietechnik (263)
- Fachbereich Medizintechnik und Technomathematik (252)
- Fachbereich Luft- und Raumfahrttechnik (213)
- Fachbereich Maschinenbau und Mechatronik (209)
- Solar-Institut Jülich (167)
- IfB - Institut für Bioengineering (155)
- Fachbereich Bauingenieurwesen (139)
- Fachbereich Wirtschaftswissenschaften (73)
- ECSM European Center for Sustainable Mobility (62)
Language
- English (1191)
- German (483)
- Italian (1)
- Multiple languages (1)
- Spanish (1)
Document Type
- Conference Proceeding (1677) (remove)
Keywords
- Biosensor (25)
- Blitzschutz (15)
- CAD (11)
- Finite-Elemente-Methode (11)
- civil engineering (11)
- Bauingenieurwesen (10)
- Lightning protection (9)
- Einspielen <Werkstoff> (6)
- Telekommunikationsmarkt (6)
- shakedown analysis (6)
The quest for life on other planets is closely connected with the search for water in liquid state. Recent discoveries of deep oceans on icy moons like Europa and Enceladus have spurred an intensive discussion about how these waters can be accessed. The challenge of this endeavor lies in the unforeseeable requirements on instrumental characteristics both with respect to the scientific and technical methods. The TRIPLE/nanoAUV initiative is aiming at developing a mission concept for exploring exo-oceans and demonstrating the achievements in an earth-analogue context, exploring the ocean under the ice shield of Antarctica and lakes like Dome-C on the Antarctic continent.
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.