The process by which humans distinguish visually between letters of the alphabet is a matter of considerable theoretical and practical interest. On the theoretical side, it is a classical problem in psychology, and has served as a test case for numerous theories of feature analysis and pattern recognition. On the practical side, it is relevant to the effective design of any textual display. Fur...

We describe an object oriented set of tools for carrying out integrated hierarchical pattern recognition in a tracking system composed of an arbitrary set of tracking devices. Two fundamental components are part of the method, a LayerSet object for pattern recognition operations and a TrackFilter object for manipulation of track candidates. The other major components in the package are: a manag...

In this paper we describe a system designed to facilitate learning of handwriting. The goal of this research is to teach visually impaired users to sign their name. Handwriting is a complex fine motor task that is best learned through doing with immediate feedback. The system described combines pen-input models with haptic trajectory playback and passive guidance through a force feedback device...

Abstract Artificial neural networks use a lot of coefficients that take great deal computing power for their adjustment, especially if deep learning are employed. However, there exist coefficients-free extremely fast indexing-based technologies work, instance, in Google search engines, genome sequencing, etc. The paper discusses the methods pattern recognition. It is shown recognition applicati...

It has been recognized for a long time that a conventional sequential processor is inefficient for operations on pictorial data where relatively simple operations need to be performed on a large number of data elements (pixels). Though many parallel processing architectures for picture processing have been proposed in the past, very few have actually been implemented due to the costs involved. ...