Achieving good text detection and recognition results for multi-script-oriented images is a
challenging task. First, we explore bit plane slicing in order to utilize the advantage of the …

Knowledge of an automobile driver's intended actions (eg, to turn, change lanes, etc.) could
facilitate the integration of intelligent vehicle systems with the driver. The actions can be …

Previous studies on temporal and tonal aspects of languages are usually based on limited
data from a small number of subjects. It is difficult to know whether these findings can really …

As robots enter the human environment and come into contact with inexperienced users,
they need to be able to interact with users in a multimodal fashion—keyboard and mouse …

This thesis is not substantially the same as any other that I have submitted for a degree or
diploma or other qualification at any other university. This thesis is entirely the result of my …

This paper presents new methods for training large neural networks for phoneme probability
estimation. A combination of the time-delay architecture and the recurrent network …

This paper presents new methods for training large neural networks for phoneme probability
estimation. An architecture combining time-delay windows and recurrent connections is …

In off-line handwriting recognition, classifiers based on hidden Markov models (HMMs) have
become very popular. However, while there exist well-established training algorithms, such …

The study of multiple classifier systems has become an area of intensive research in pattern
recognition. Also in handwriting, recognition, systems combining several classifiers have …

Presentation failures in a website can negatively impact end users' perception of the quality
of the website, the services it delivers, and the branding a company is trying to achieve …