Design and implementation of a real-time vision-assisted system for micro-scaled object identification and tracking

A prototypical vision system together with a novel, adaptive image processing method that identify and track multiple moving micro-scaled objects in real time are developed and discussed in this paper. The system and method proposed and discussed in this paper can potentially provide a reasonable low cost means to satisfy operational needs of utilizing conventional microscopes in micro-manufacturing, assembling, and bio-medical applications. With the proposed system and method, it is demonstrated that multiple sub-micron scaled moving objects can be identified and traced unambiguously. The trajectory database resulted from the system has implications in integration with micro actuators for micro-servoing in the mentioned high precision engineering applications.