000 Allgemeines, Wissenschaft
Refine
Document Type
- Bachelor Thesis (1)
- Book (1)
- Master's Thesis (1)
- Researchpaper (1)
Language
- English (4) (remove)
Has Fulltext
- yes (4)
Is part of the Bibliography
- no (4)
Keywords
- Convolutional Neural Network (1)
- Debugging (1)
- Deep Learning (1)
- Distributed Systems (1)
- Fully Convolutional Network (1)
- JBoss (1)
- Java (1)
- Java EE (1)
- Java Enterprise (1)
- Kultur (1)
Institute
In recent years new trends such as industry 4.0 boosted the research and
development in the field of autonomous systems and robotics. Robots collaborate and
even take over complete tasks of humans. But the high degree of automation requires
high reliability even in complex and changing environments. Those challenging
conditions make it hard to rely on static models of the real world. In addition to
adaptable maps, mobile robots require a local and current understanding of the scene.
The Bosch Start-Up Company is developing robots for intra-logistic systems, which
could highly benefit from such a detailed scene understanding. The aim of this work
is to research and develop such a system for warehouse environments. While the
possible field of application is in general very broad, this work will focus on the
detection and localization of warehouse specific objects such as palettes.
In order to provide a meaningful perception of the surrounding a RGB-D camera is
used. A pre-trained convolutional network extracts scene understanding in the form
of pixelwise class labels. As this convolutional network is the core of the application,
this work focuses on different network set-ups and learning strategies. One difficulty
was the lack of annotated training data. Since the creation of densely labeled images
is a very time consuming process it was important to elaborate on good alternatives.
One interesting finding was that it’s possible to transfer learning to a high extent from
similar models pre-trained on thousands of RGB-images. This is done by selective
interventions on the net parameters. By ensuring a good initialization it’s possible
to train towards a well performing model within few iterations. In this way it’s
possible to train even branched nets at once. This can also be achieved by including
certain normalization steps. Another important aspect was to find a suitable way
to incorporate depth-information. How to fuse depth into the existing model? By
providing the height over ground as an additional feature the segmentation accuracy
was further improved while keeping the extra computational costs low.
Finally the segmentation maps are refined by a conditional random field. The joint
training of both parts results in accurate object segmentations comparable to recently
published state-of-the-art models.
When searching for bugs in Java enterprise applications, an essential part of the
eort consists in redeploying the source code and relaunching the server over and
over. In order to improve this situation, this thesis suggests the implementation
of a runtime debugging tool. The tool's purpose is to facilitate the enrichment of
operating application code with logging statements, which are inteded to generate
additional output concerning the webapp's current state. On behalf of this
so-called instrumentation, the actual process of debugging could be supported
and accelerated without having to interrupt the server's execution.
Due to the signicance of Java EE as well as Spring for today's enterprise development,
the implementation of a dedicated debugging tool for each platform
shall be covered. Both solutions pursue the same goal, but dier in the approach
and the programming paradigm forming their basis. This document introduces
their implementation details and evaluates them against a specication that de-
nes the general conditions and expectations in terms of the capabilities of a
satisfying result.
Talking about highly scalable and reliable sys-
tems, issues like logging and monitoring are often
disregarded. However, being able to manage to-
day’s software systems absolutely requires deep
knowledge about the current state of applications
as well as the underlying infrastructure. Extract-
ing and preparing debug information as well as
various metrics in a fast and clearly arranged
manner is an essential precondition in order to
handle this task.
Since we at Bertsch Innovation GmbH also
face increasing requirements concerning Media-
Cockpit as one of our core products, we decided
to establish a centralized logging infrastructure
in order to come up to the application’s evolution
towards a more and more distributed system.
In this paper, I want to describe the steps
that I have taken in order to setup a functioning
logging tool stack consisting of Elasticsearch,
Logstash and Kibana (usually abbreviated as ELK stack ). Besides outlining proper
setup and configuration, I will also discuss possi-
ble pitfalls as well as custom adjustments made
when ELK did not meet our demands.
Free Culture : how big media uses technology and the law to lock down culture and control creativity
(2004)
The struggle that rages just now centers on two ideas: piracy and property. My aim in this book s next two parts is to explore these two ideas. My method is not the usual method of an academic. I don t want to plunge you into a complex argument, buttressed with references to obscure French theorists however natural that is for the weird sort we academics have become. Instead I begin in each part with a collection of stories that set a context within which these apparently simple ideas can be more fully understood. The two sections set up the core claim of this book: that while the Internet has indeed produced something fantastic and new, our government, pushed by big media to respond to this something new, is destroying something very old.Rather than understanding the changes the Internet might permit, and rather than taking time to let common sense resolve how best to respond, we are allowing those most threatened by the changes to use their power to change the law and more importantly, to use their power to change something fundamental about who we have always been. We allow this, I believe, not because it is right, and not because most of us really believe in these changes.We allow it because the interests most threatened are among the most powerful players in our depressingly compromised process of making law. This book is the story of one more consequence of this form of corruption a consequence to which most of us remain oblivious.