Eye Movement-Based User Interaction Behavior Analysis Technology

As a new type of interaction method,eye-movement interaction has been widely used in many fields such as smart driving,smart medical care,smart home,and smart education because of its initiative,openness,intelligence,and immediacy.Human vision is bidirectional.When acquiring or outputting information,the eye movement information contains implicit and explicit interaction intentions.The information plays an important role in the eye movement interaction technology.By studying the rules of eye movement behavior,effectively predicting the user's implicit interaction intention and reasonably expressing the explicit manipulation intention is the key to improving the intelligence and naturalness of the eye movement interaction system.Therefore,this thesis studies the eye movement behavior in the interaction process from two aspects: sight feedback and sight control.In terms of sight feedback,this thesis implements an eye-movement interest area extraction algorithm based on eye movement information.After that,the attention degree of gaze behavior is described from two dimensions of time and space,and the fuzzy kernel clustering algorithm with visual characteristics is used to softly divide the gaze point.Then,an attention calculation model fused with eye movement interest areas is constructed to analyze the inherent rules of potential interaction intentions and eye movement behaviors,thereby effectively predicting implicit intentions in the user interaction process.Finally,the effectiveness of the model is verified by experiments.In terms of gaze control,this thesis builds an eye movement interaction evaluation index system in human-machine collaboration.After that,an eye movement behavior library is constructed based on eye movement behavior characteristics and region-based eye movement interaction mechanisms,and a recognition algorithm for eye movement behavior is designed and implemented.Then,taking the eye movement control task of the UAV(unmanned aerial vehicle)as the background,the interactive behavior data is obtained through interactive experiments and the rules are analyzed.Finally,the eye movement behavior instructions that conform to the user's behavior habits are obtained for the control instructions of the UAV.Meanwhile,the effectiveness of the eye movement behavior recognition algorithm is verified in experiments.