Gesture Design And Decision-making Optimization Algorithm In "contact + Non-contact" Interaction

With the rapid development of computer software and hardware technology,people's demand for product experience is getting higher.Natural interaction has become the mainstream trend of human-computer interaction technology development.Gesture input in the "contact +non-contact" interaction enhances the computer's ability to collect information and makes the interaction closer to natural interaction.The research on "contact + non-contact" gestures design and optimization is of great significance for improving related application systems efficiency.The "contact + non-contact" multi-interaction model is currently in the conceptual research stage.According to the relevant research of the gesture principle,the key problems in designing gestures and optimization algorithm for "contact + non-contact" interactions are:1.There are cross-repetition and differences in the principles of gesture design that are emphasized in each literature.How to collect the principles of gesture design and build a systematic set of gesture principles is a major challenge.2.Currently,there are not enough "contact + non-contact"gestures for design reference and we need to reduce the influence of human factors in user feedback information.How to deal with the collected user feedback information effectively and make a reasonable decision on gesture improvement direction is a major challenge.3.How to improve the existing "contact + non-contact" gesture design according to the direction of gesture improvement and obtain a gesture with better usability is a major difficulty.Aiming at these key problems,we research on how to build systematic gesture design principles to effectively process the collected user feedback information and make a reasonable decision on the gestures improvement direction.The following results have been achieved:(1)Build systematic gestures design principles:The existing gesture design principles are not systematic.According to the difference of gesture principles' application scope,the design principles of gesture are analyzed from three angles:human,computer and interaction process.We propose a tree structure to organize the principle set of gestures.It provides a foothold for gesture design in the gesture interaction system and provides a comprehensive evaluation criterion for the designed gestures.(2)propose an optimization algorithm of "contact + non-contact" gestures:quantified indicators are studied to guide the decision on gestures improvement direction.Taking gesture interaction design as the support attributes of gestures,proposed a decision algorithm for the gesture improvement direction based on the importance of attributes.Calculate the global weight for each attribute of the specified gesture according to the Fuzzy Analytic Hierarchy Process(FAHP)algorithm.Calculate the local weight for each attribute of the specified gesture according to the user feedback.The global weights and the local weights are harmonized into comprehensive weights vector.According to the user's comprehensive impression on the specified gestures,each attribute's score is divided into two classes and processed separately.According to the positive and negative influence degree of each attribute,we obtained the weight distribution of each attribute in the weight vector.Each score of the attributes is calculated by the standardized score bias respectively.The rank of improvement necessity on each attribute is calculated by the weight distribution and the standardized score bias.It provides a stable method to optimize the gesture decision,which ensures the continuous optimization of gestures.(3)Design a set of multi-gestures for touch gestures and non-contact gestures:Based on the direction of gesture improvement found by the algorithm mentioned above,we improve the existing "contact + non-contact" gesture design to get better practical gesture.The experimental results show that the support rate of the gesture based on this algorithm is 25%higher than that based on questionnaires.