Fuzzy posture input for virtual-hand modals

Virtual-hand input realizes the natural and dexterous functionality in dire ct human-computer interaction. The recognition of virtual-hand models is di fficult due to two reasons: the complexity of hand structure and the lack o f accurate measure. which may be caused by either mechanical noise or human factors. This paper presents a novel fuzzy-logic recognition system that c an effectively deal with imprecise hand data The system consists of three c omponents: the classifier, identifier, and posture database. Fuzzy-logic pr ocessing is applied in both the classifier (to build class-indexing structu re in the posture database) and in the identifier (to find the most likely match from the database in real-time VR applications), The posture database provides a GUI interface for the user to browse the posture images and int eractively update the database by adding and deleting the sample postures, as well as adjusting the certainty threshold in recognition. Our experiment s show that the fuzzy-logic method can keep nearly constant performance in recognition using tens of microseconds, even as the size of posture databas e increases. The recognition late declines only slightly when the standard derivation of the noise distribution (Gaussian distribution) in the input p arameters is below 15 deg. The bounded value in uniform distribution to ach ieve accurate recognition is below 20. The results show that the fuzzy-logi c processing can improve the tolerance of noise or imprecise data in an eff icient way. A free-hand modeler based on fuzzy-logic processing has been de veloped for creating virtual objects and Web-based 3-D VRML worlds.