Hence, semantic radical information is useful, but insufficient, for character recognition.
Research and development work will continue on still more advanced systems, including the possible use of optical character recognition techniques for the automatic sorting of certain types of post-coded mail.
Pencept employed a proprietary technology for on-line character recognition, based on a functional attribute model of human reading.
As a result, character recognition primarily depends on parallel information retrievals at multiple representational levels rather than intraword analysis and information integration.
There are also applications to straightforward image recognition, including some in medicine as well as the familiar one of handwritten character recognition.
His current work is developing speech and language front-ends to operational neural network systems for optical character recognition and vehicle tracking.
In all probability, competent logographic readers progressively become aware of these constraints and develop the capacity to utilize them in accelerating character recognition.
A more mundane example is the problem of character recognition.
