Research And Realization For The Image Processing System Of Pharmacy Robot

For the past20years, the subjects such as microelectronics, communication, computer,artificial intelligence, control and image processing have developed rapidly. The robot that isdeveloped based on automation technology and computer technology, combining variousmodern information technology have been widely applied in the industrial manufacturingfield Meanwhile, robots start to play a unique role in medical field, especially in the aspectsof surgery rehabilitation and medical care robots. Dispensing robot is a kind of medical carerobots, among all of the many systems of it, this article chooses to do some research anddevelopment on the image processing subsystem of the dispensing robot.The function of the image processing system of the dispensing robot is to acquire thethree-dimensional coordinate of the medicine bottle’s bottleneck while it conducts the imagecollection of medicine bottles. We divide it into two parts: the human-computer interactionand the image processing.In the human-computer interaction, the MFC (Microsoft Foundation Class) in the VisualC++IDE (integrated development environment) and Microsoft launched by Microsoftapplication framework is used for the development in this article. Via the completehuman-computer interaction interface, a user can make dispensing robots start or pause theirwork, and export the images which the user is interested in and place them on file.In the image processing part of the image processing system, the article chooses OpenCVto achieve its goal. OpenCV is the Open Computer Vision of Intel which is edited in C++language and whose function execution speed is fast, and can be applied to the real-timeprocessing of images, and can run on Linux/Windows and other operating systems.In this article, firstly, camera calibration and location calibration for the surface of thecamera and medicine bottle are conducted in order to obtain the parameters required by theSystem. Secondly, the image preprocessing, image edge detection, bottleneck point extractionof the medicine bottle in the image are conducted on the image obtained by the camera to findthe actual place of the bottleneck point of the medicine bottle in the image. Finally,combining the results of the above, we find the actual three-dimensional coordinate of themedicine bottle’s bottleneck point.