From the viewpoint of data analysis, the public spaces are in flow state because passengers in there are moving and transient. In order to recognize the state of public spaces, it is essentially important to catch up continuous changes of moving people. It is desirable to establish a method to deliver information according to such transient and continuously changing states. Thus, we propose a mass's context-aware digital signage system to be applied to the automatic information delivery for people in the public spaces. Our system utilizes sensors installed in public facilities to obtain statistical data for representing people movements. And then, the system classifies the obtained statistical data, within a specific time-window, into several predefined categories corresponding to advertisement criterion. The most important feature of this system that it controls the 3-axis of digital signage according to the recognized status of people. This 3-axis consists of 2-dimentional(X and Y)layout structure of the digital signage, and the update frequency of each content on the digital signage display. This system allows for controlling size, place, and frequency of showing contents, in a quantitative, fine-grained, and context-aware way. By using this system, it is possible to improve the performance of digital signage because this system offers different contents for people in the public spaces. For example, students, DINKS, couples with children, and seniors will look different contents even if there are no personalization function, by analyzing the structure of the person of the place. This paper shows a prototype implementation using HTML5 technology of digital signage system that we propose, and examines the effectiveness of controlling quantitatively the screen of the digital signage that is installed in a public spaces.
|