A window on the screen would display the live video stream of the camera. The data in turn would be interpreted by Arduino and rotate the camera. On a change of one of the variables in GC, this would send data over that channel. function was written in C# that would create a communication channel with the serial port. The software implementation is depicted in Fig. With the integration of openCV (OpenCV, 2009) for face recognition and tracking, the background image is ignored and the performance of the motion tracking is improved. Therefore a delay was added between instructions, and it was found that motion tracking of a single object along an evenly coloured background worked best. When the camera moves however, consecutive images look completely different. The largest area of motion in the picture indicated the point of interest and the software would make the camera point towards that point of interest. The software was adapted to send data to the serial port in a format that Arduino could interpret. Therefore the software can tell that firstly there is motion in the image, and secondly where in the image that motion takes place.
Areas in the image that are different from the previous are marked. Open source software (openFrameworks, 2009) was used and adapted to control the webcam and compare consecutive frames. The webcam was also connected to a computer. If the signal comes in the right format, the signal is decoded and the two servos are instructed to rotate to a defined angle. The controller board was programmed to wait for a signal on the serial port. The board can also stay connected to a USB port of a computer and communicate through a serial protocol. Arduino controller boards connect to other hardware and have a chip that can be programmed and then run in a stand-alone mode. Arduino is an open-source electronics prototyping platform based on flexible, easy-to-use hardware and software (Arduino, 2009).
The two servos are controlled using an Arduino controller board. The view of the camera could therefore cover roughly half a sphere like the eye of a chameleon (Fig. The orientation of one servo motor allows the camera to rotate along a vertical axis, and the other allows for movement along a horizontal axis. The physical model of the eye was a little robot that consists of a (web) camera and two servo motors. The virtual model mimics the physical model and vice versa, thereby allowing for calibration and for trans-reality feedback loops. The experiment setup was a simple relation between a physical model and a virtual representation of an eye. It allows for input into GC and output from GC. The second experiment in SmartGeometry 2009 is using a different protocol. data when the device’s states have substantially changed.
0 kommentar(er)
