Abstract
We are developing a portable intelligent medicine case to assist users in maintaining their good medication compliance. The medicine case is composed of a smartphone with a camera and a smartphone cover which has three storage compartments for medicine. The cover also has one plane mirror and three cylindrical mirrors. These cylindrical mirrors enable the camera to capture all the compartments by reflection. In this paper, we propose an image processing technique for object detection. This technique takes advantage of the shapes of the cylindrical mirrors and calculates coefficients of three approximated curves corresponding to the three mirrors. We confirmed that the medicine case utilizing the proposed technique was able to recognize the presence of medicines in each compartment through experimentation.
