A system for cursive handwriting recognition based on human reading models

Computer technology has improved rapidly over the years, and with more powerful machines becoming ever smaller and cheaper; there is no indication of the trend slowing. The latest growth area is in portable personal computers providing powerful facilities to the mobile user. Alongside this development has been improvements to the human computer interface (HCI) leading to the modern notebook computer. Recent developments in the HCI, especially the development of interfaces and support software based on a stylus have led to the concept of pen-computers. These can fully exploit this mobility as all user-computer interaction is conducted on the computer´s display using a pen rather than a keyboard. The resulting machine could in principle be used with no more difficulty than a pen and paper. The major area of technological development required for such a device to become practicable is in the area of handwriting recognition as this is the obvious and necessary method for entering data onto such a device. Cursive script recognition (CSR) research aims to comprehend unconstrained, natural handwriting