Vision Club of Finland awards Vuokko Vuori with Vision Prize 98 

VCF has annually awarded the best thesis in the area of MV technology research and application. The received applications were concrete and practical and included industrial potential. This year the winner is Vuokko Vuori, who made meritorious research in adaptive real-time recognition of handwriting and published the results in her Master of Science thesis. Congratulations to the winner of 3000 FIM award and many thanks to all of them, who applied to the competition. 
 

Adaptation in on-line recognition of handwriting 

Due to the increasing popularity of hand-held computers, digital notebooks, and advanced cellular phones, automatic on-line recognition of hand-written text has been gaining more interest lately. The traditional methods of man-machine communication, a keyboard and pointing devises, are very inconvenient to use and expensive to implement when the size of the machine is only slightly bigger or same size as human palm. Therefore, handwriting recognition is a very attractive input method.  

The most prominent problem in handwriting recognition is the vast variation in personal writing styles. A recognition system should be insensitive with minor variations and still be able to distinguish different but sometimes very similar looking characters. Recognition systems should, at least in the beginning, be able to recognise many writing styles. Such multi-user systems usually have problems with recognition accuracy. One way to increase performance is adaptation, which means that the system learns its user's personal writing style.  

The main goals of the master's thesis Adaptation in On-line Recognition of Handwriting were to make a literature survey of the state-of-art recognition methods and to develop adaptive methods for on-line recognition of isolated hand-written characters. The experiments reported in the work showed that with the suggested adaptation methods, recognition accuracy high enough to be acceptable for a real-world application could be attained for most of the writers. 

The work described in the thesis has been carried out in co-operation with Nokia Research Center. The Technology Development Centre Finland (TEKES) as part of the technology programme Adaptive and Intelligent Systems Applications have funded it.