| Mobile and wearable computing have becoming more prevalent in our daily lives in these ten years. Especially, with the growing popularity of various smart phones, ultra mobile PCs, tablet PCs and head-mounted displays (HMD), increasing emphasis has been placed on their human computer interaction (HCI) aspect. Nowadays, the output of the HCI of mobile and wearable computing is mainly based on hand held small display (HSD) and HMD. The input of the HCI is mainly based on the pointing interaction, eg. text and cursor pointing input. However, the restricted display field of view (FOV) devices which been adopted in the HCI, would result in some serious problems. Such as various target size (both encounter with small sized and common sized targets during the interaction), as well as various display FOV (FOV is restricted but is commonly varied). Then, those pointing based HCI performance would be significantly affected, and further theoretical difficulty would be resulted in. In addition, the usability, which added from the HMD effi-ciency evaluation and the HMD/HSD GUI adjustment, would also influence the usability of mobile and wearable computing systems. Thus, targeting at those problems, both theo-retical and practical implications would be strengthened by these work, eg. HCI efficiency analysis, new HCI model design, and interaction (evaluation) method improvement.This dissertation is mainly focuses on the models and key issues for display FOV restricted HCI in the context of mobile and wearable computing. The dissertation was founded by the National High Technology Research and Development Program (863 Pro-gram) "An ultra light weight wide field of view see-through head-mounted display based on free form surface prim" (ID:2009AA01Z310), and the "The architecture and key tech-nologies of special wearable network terminal" (ID:2006AA01Z113). The main contri-butions are summarized as follows:1. A new HCI model for various target size conditions was proposed. The new model would provide a theoretical support and a research tool, for the HCI under the same restricted display FOV but various target size conditions:The poor usability of the traditional Fitts'model under various target size conditions has been proved; A new index of difficulty and a new physical model for various target size interaction have been proposed, on the basis of Shannon information theory and kinematics structure of human; The quantitative efficiency of pointing task has been investigated under the various target size conditions, and the applicable scope of the new model has also been proposed; The evaluation results show that, the new model could be used both as a good descriptive model and a good predictive model, under various target size target conditions.2. A new HCI model for various display FOV conditions was proposed. The new model would have implications for providing a theoretical support and a research tool, which under various restricted display FOV but same target size conditions:The influence of various display FOV on HCI efficiency was proved significant, and the usability of traditional Fitts'model under various display FOV conditions was proved low; Then, on the basis of Shannon information theory and human factor mechanisms, a new index of difficulty and a new physical model for various display FOV conditions were then proposed; The quantitative efficiency of the interaction has been investigated under a wide range display FOV conditions and various tasks; The evaluation results showed that the new model could be used both as a good descriptive and as a good predictive model under the various display FOV conditions.3. A new evaluation method for HMD's efficiency was proposed, on the basis of user in-teraction study and our various target size interaction model. The new method would solve the difficulty when applying the traditional method, which requires professional instrument, software and background knowledge. Thus it could easily be applied by most of the end users:The relationship between display impulse and human machine system has been analyzed on the basis of several human mathematical models; The factors, which affect a HMD's comprehensive and imagery performance, were clas-sified; Three steps and seven rules have also been proposed for conducting a contrast user study; The detailed quantitative relationships targeting at three HMDs have been measured by a user study; Then, two HMD efficiency evaluation methods were pro-posed. The first one based on the user study and the cross-over-model method, the second one on the basis of the user study and the various target size model method; The evaluation result from the traditional method showed that, both of our new meth- ods could be used for evaluating the comprehensive and imagery performance of the HMDs. Additionally, in contrary to the traditional method, our new methods, espe-cially the second one, have a better usability;4. Two self-adapting graphical user interface (GUI) scaling methods, which targeting at HMD and HSD, have been proposed on the basis of our various display FOV model. On one hand, our new methods eliminate the negative influence on the HCI perfor-mance, which results from the physical properties of the device and the interaction status of the user. On the other hand, our new methods also solve the problem that the traditional GUI scaling method all involves in manually manipulation:Two GUI scaling methods, which targeting at HMD and HSD, were proposed on the basis of border conditions of our various display FOV model; The HMD GUI scaling method has been implemented into a custom HMD and a custom X86 computer mainboard by the EEDID information; In addition, the HSD GUI scaling method was implemented on an ultra mobile PC by means of Adaboost human eyes tracking; Then, two evalu-ation procedures were conducted for both of the scaling methods. The results showed that, as for those FOV restricted devices, using the scaling methods could benefit the pointing task performance. |
PDF Full Text Request