Hanover University of Applied Sciences

www.fhdw-hannover.de
Hanover, Germany
SEARCH FILTERS
Time filter
Source Type

Klinger V.,Hanover University of Applied Sciences
3rd International Workshop on Innovative Simulation for Health Care, IWISH 2014 | Year: 2014

The use of nerve signals for a prosthesis control or limb stimulation is one great challenge in medical technology. It requires a recording of the electroneurogram (ENG) data and an identification of the motion-based action potentials of motoric, feedback and sensoric nerves within the corresponding neural bundle. We have realized a prototyping system for the data acquisition of ENG data, including an identification framework, described in (Klinger and Klauke 2013). In this paper we introduce the verification concept of the identification process using synthetic datasets generated based on robot manipulator and electroneurogram simulator. The objective is to define a method to compare motion based trajectories and their corresponding ENG signals to prepare the future analysation and identification of human ENG data.


Lowe M.,Hanover University of Applied Sciences
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) | Year: 2012

There are three major algebraic approaches to graph transformation, namely the double-pushout (DPO), single-pushout (SPO), and sesqui-pushout approach (SqPO). In this paper, we present a framework that generalises all three approaches. The central issue is a gluing construction, which is a generalisation of the construction introduced in [14]. It has pushout-like properties wrt. composition and decomposition, which allow to reestablish major parts of the theory for the algebraic approaches on a general level. We investigate parallel independence here. © 2012 Springer-Verlag.


Steinke K.-H.,Hanover University of Applied Sciences
CSAE 2012 - Proceedings, 2012 IEEE International Conference on Computer Science and Automation Engineering | Year: 2012

In many Botanical Museums in the world exist millions of dried plants on paper sheets. They were collected in the last two hundred years. The collectors left annotations about the plants on the sheets. Some of them are written in handwriting others by a typewriter. There exist also comments on printed forms which are glued on the sheets. By an automatic OCR of the scanned herbarium objects the recognized texts can be put into a database. Commercial OCR-programs like Finereader or Omnipage are capable to recognize undisturbed printed texts in a correct way. Unfortunately we deal with historical material which often contains overwritten or crossed out words, the paper is yellowed, molded and has other artifacts. Moreover it seems to be extremely complicated to detect text regions in a complex environment with objects like roots, leaves, stamps, bar codes, yardsticks, color charts etc. The old writing leads to read errors which are not the biggest problem, because they can be compensated by tolerant database queries. Worse is that sometimes text regions cannot be localized. In this paper a method is presented which helps Omnipage to detect missed text and also can distinguish printed text from handwriting. © 2012 IEEE.


Lowe M.,Hanover University of Applied Sciences
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) | Year: 2015

The paper extends Sesqui-Pushout Graph Rewriting (SqPO) by polymorphism, a key concept in object-oriented design. For this purpose, the necessary theory for rule composition and decomposition is elaborated on an abstract categorical level. The results are applied to model rule extension and type dependent rule application. This extension mechanism qualifies SqPO – with its very useful copy mechanism for unknown contexts – as a modelling technique for extendable frameworks. Therefore, it contributes to the applicability of SqPO in software engineering. A version management example demonstrates the practical applicability of the combination of context-copying and polymorphism. © Springer International Publishing Switzerland 2015.


Stein M.,Hanover University of Applied Sciences
AIP Conference Proceedings | Year: 2011

Let Fn and Ln denote the Fibonacci and Lucas numbers, respectively. D. Duverney, Ke. Nishioka, Ku. Nishioka and I. Shiokawa proved that the values of the Fibonacci zeta function ζF(2s)= Σn=1 ∞Fn -2s are transcendental for any s∈ℕ using Nesterenko's theorem on Ramanujan functions P(q), Q(q), and R(q). They obtained similar results for the Lucas zeta function ζL(2s)=Σn=1 ∞L n -2s and some related series. Later, C. Elsner, S. Shimomura and I. Shiokawa found conditions for the algebraic independence of these series. In my PhD thesis I generalized their approach and treated the following problem: We investigate all subsets of {n=1Σ∞1/F n 2s1,n=1Σ∞(-1)n+1/F n 2s2,n=1Σ∞1/Ln 2s3,n= 1Σ∞(-1)n+1/Ln 2s4:s 1,s2,s3,s4∈ℕ} and decide on their algebraic independence over ℚ. Actually this is a special case of a more general theorem for reciprocal sums of binary recurrent sequences. © 2011 American Institute of Physics.


Mund B.,Hanover University of Applied Sciences | Steinke K.-H.,Hanover University of Applied Sciences
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) | Year: 2010

About 3.5 million dried plants on paper sheets are deposited in the Botanical Museum Berlin in Germany. Frequently they have handwritten annotations (see figure 1). So a procedure had to be developed in order to process the handwriting on the sheet. In the present work an approach tries to identify the writer by handwritten words and to read handwritten keywords. Therefore the word is cut out and transformed into a 6-dimensional time series and compared e.g. by means of DTW-method. A recognition rate of 98.6% is achieved with 12 different words (1200 samples). All herbar documents contain several printed tokens which indicate more information about the plant. With the token it is possible to get information who has found this plant, where this plant was found (country and sometimes the town), what kind of plant it is and so on. By using the local connections of the text it is possible to get more information from the herbar document, e.g. to find and recognize handwritten text in a defined area. © 2010 Springer-Verlag Berlin Heidelberg.


Endres H.-J.,Hanover University of Applied Sciences
Kunststoffe International | Year: 2011

The sale of plastic carrier bags made from petrochemical raw material has been banned in Italy. With continuously growing quantities of biopolymers in the marketplace, it is increasingly necessary and sensible also to consider all other alternative disposal options for biopolymer packaging alongside composting. In manufacturing and processing biopolymers fundamentally the same technology can be used as for conventional plastics. Barrier properties are one area where biopolymers with their polar structure show a significant difference when compared to conventional polymers like polyethylene, polypropylene or polyethylene terephthalate. Amongst the biogdegradable biopolymers only PLA, the biopolymer that has been furthest developed as a packaging material up until now, and polyhydroxy alkanoates show a certain degree of barrier functionality for water vapor. The most promising possibility to improve the barrier properties of biopolymers is said to be coating.


Lowe M.,Hanover University of Applied Sciences
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) | Year: 2016

Sesqui-pushout rewriting is an algebraic graph transforma- tion approach that provides mechanisms for vertex cloning. If a vertex gets cloned, the original and the copy obtain the same context, i.e. all incoming and outgoing edges of the original are copied as well. This behaviour is not satisfactory in practical examples which require more control over the context cloning process. In this paper, we provide such a control mechanism by allowing each transformation rule to refine the underlying type graph. We discuss the relation to the existing approaches to controlled sesqui-pushout vertex cloning, elaborate a basic theoretical framework, and demonstrate its applicability by a practical example. © Springer International Publishing Switzerland 2016.


Klinger V.,Hanover University of Applied Sciences
4th International Workshop on Innovative Simulation for Health Care, IWISH 2015 | Year: 2015

Simulation, modelling and verification are powerful methods in computer aided therapy, rehabilitation monitoring, identification and control. They are major prerequisites to face great challenges in medical technology. To realize tasks and services like an on-line data monitoring or a nerve signal based prosthesis control, smart, intelligent and mobile systems are required. Here we present data acquisition and learning systems providing methods and techniques to acquire electromyogram (EMG)-and electroneurogram (ENG)-based data for the evaluation and identification of biosignals. We focus on the development, integration and verification of platform technologies which support this entire data processing. Simulation and verification approaches are integrated to evaluate causal relationships between physiological and bioinformatics processes. Based on this we are stepping up efforts to develop substitute methods and computer-aided simulation models with the objective of reducing experiments on animals. This work continues the former work about system identification and biosignal acquisition and verification systems presented in (Bohlmann, Klinger, and Szczerbicka 2010; Klinger and Klauke 2013; Klinger 2014). This paper focuses on the next generation of an embedded data acquisition and identification system and its flexible platform architecture. We present results of the enhanced closed-loop verification approach and of the signal quality using the Cuff-electrode-based ENG-data acquisition system.


Wolter U.,University of Bergen | Konig H.,Hanover University of Applied Sciences
Applied Categorical Structures | Year: 2015

Reliable semantics for software systems has to follow the semantics-as-instance principal (fibred semantics) rather than the semantics-as-interpretation principal (indexed semantics). While amalgamation of interpretations is simple and nearly always possible, amalgamation of instances is very much involved and not possible in many cases. A condition when two compatible instances (a span of pullbacks) are amalgamable, is presented for presheaves, i.e. functor categories SET. Based on this individual condition we prove further a total condition for amalgamation which simultaneously yields a necessary and sufficient condition for pushouts to be Van Kampen squares. As a necessary and adequate basis to achieve these results we provide a full revision and adaption of the theory of descent data in topoi for applications in diagrammatic specifications including graph transformations. Especially, we characterize Van Kampen squares in arbitrary topoi by pullbacks of categories of descent data. © 2013, Springer Science+Business Media Dordrecht.

Loading Hanover University of Applied Sciences collaborators
Loading Hanover University of Applied Sciences collaborators