Pervasive Image Computation: A Mobile Phone Application for getting Information of the Images
This addresses the need for multimedia-based information processing systems for mobile users, but it appears incremental as it builds on existing concepts of image processing and database integration.
The paper tackles the problem of multimedia information processing by proposing a pervasive image computation system where users take pictures with mobile phones to obtain information about objects. The implementation demonstrates two flexible architectures (online and offline) that can be extended to more complex pervasive multimedia systems.
Although many of the information processing systems are text-based, much of the information in the real life is generally multimedia objects, so there is a need to define and standardize the frame works for multimedia-based information processing systems. In this paper we consider the application of such a system namely pervasive image computation system, in which the user uses the cellphone for taking the picture of the objects, and he wants to get some information about them. We have implemented two architectures, the first one, called online architecture, which the user sends the picture to the server and server sends the picture information directly back to him. In the second one, which is called offline architecture, the user uploads the image in one public image database such as Flickr and sends the ID of the image in this database to the server. The server processes the image and adds the information of the image in the database, and finally the user can connect to the database and download the image information. The implementation results show that these architectures are very flexible and could be easily extended to be used in more complicated pervasive multimedia systems.