Abstract:This paper researches a new kind of human computer interaction model: It can operate the moving of the mouse in 2D (PC desk) by hand movement in 3D.The new sort of human computer interaction model makes the cell phone become a new tool for human computer interaction, instead of the original tradition ones. We present the method that the cell phone which based on the ARM platform and Symbian system detect a mobile phone user’s hand movement direction in real time or detect the only object movement direction by analyzing image sequences captured by built-in camera on different movement background. So as that, we carried on an investigation on movement and motionless background respectively:
1. Movement background. It makes the cell phone in the hand to detect the hand movement by the movement of the cell phone. The algorithm relies on the change of the background.
2. Motionless background. It means that the cell phone is in still state and the moving direction is detected by the only object movement.
In the process of image treating, We use three algorithms, i.e. the optimization algorithm、improving algorithm1 and improving algorithm2.We also perform some relative experiment for analyzing performance of these algorithms. As a result, the conclusion of experiment increases the precision to a certain extent.
This research takes advantage of the resource on the cell phone itself and the interactive operation with human, which improves the performance of human computer interaction on movement space. It can be used successfully in a number of areas :such as working、entertainment、leisure and so on.
Keywords: Mobile Devices , Computer Vision , Camera Mobile Phones , Digital Image Processing , Motion Estimation , Human Computer Interaction , Gesture Recognition