Smart device system to manage virtual information in small scale settings using augmented reality interfaces

Abstract

Smart devices have become more commonplace in the past years, surpassing mobile phones in use. Partially, this rise can be attributed to the possibility of customizing the devices with new software, which in turn are easier to create. These applications exists for many purposes, allowing new ways for the users to create and share their experiences but also provide useful services, such as GPS enabled navigation. Augmented reality (AR), meaning the addition of virtual information to the real world in a seamless manner, can be explored with the devices as well. AR requires a combination of specific hardware, that are found in compact smart devices, to be functional. Smart devices, with their large displays and cameras allow vision based detection and visual presentation of augmentations. AR applications enable smart devices to function as a “magnifying glass” to the underlying data of the observed objects or locations.

This thesis explores the new paradigm of interaction that AR allows on smart devices. This is achieved through the construction of an Android smart device application, utilizing free or open source software development tools. The main application concept features linking of user specified instances, objects or locations, to information that the user associates with the instances. By using the application, the user can create, modify and detect the added information from their environment.

The research was initiated by exploring the literature related to the research domain and performing a review of the existing AR applications available on Android devices. The knowledge from the existing literature and reviews was used to formalize the research question and to define the thesis’s limitations. As the creation of a new application was the principal target of this thesis, Systems Development Research (SDR), a research approach relating and contributing to Design Science (DS), was chosen to be utilized.

The application was designed with the use of scenarios, that detailed the key features that the system should implement. The scenarios were used in the evaluation of the application as well, by identifying the deviation from the design that had occurred during the implementation. Volunteers also used the scenarios to evaluate how well the key features were implemented within the application. The applications performance and stability were also evaluated. The results of the evaluation show that the application implements the features described in the scenarios, with volunteers stressing the importance of feedback from the application to improve its use. The application is also shown to be stable and developed conforming to the design guidelines and requirements expected of it.

The creation of a AR smart device application using vision based detection requires identification of the systems requirements and the capabilities of the software development tools. For vision based detection, OpenCV offers comprehensive and cross-platform capable libraries, which can be used to implement the required AR functions.