Active imaging of micro objects with variable view

Abstract

With the development of the micro-electro-mechanical system (MEMS) and biotechnology, there is a growing interest in observation of dynamic targets in three-dimensional (3D) space. Keeping the region of interest (ROI) on the moving target from the best view direction is critical for observation. Especially for the cases of microassembly and micromanipulation, the two-directional interaction between a micromanipulator/microrobot and a vision system is an important factor for the success of these applications. Unfortunately, the conventional microscopes used in most of these applications cannot meet this requirement. Insufficient vision information in these systems, such as occlusion, a small field of view (FOV) and low depth resolution often hinders the success of micro inspection, microassembly and/or micromanipulation. In this thesis, an optomechatronic system that has been termed a Variable View Imaging System (VVIS) is introduced. The original idea of the proposed system comes from both robotics and optics. The proposed system can interact with the environment by changing optical system parameters such as the view position and orientation (pose) with four degrees-of-freedom in a compact manner. Double wedge prisms are applied in the system to steer the view angle. Different active components are integrated into one system to achieve new functions and to improve performance. The proposed system consists of a pair of wedge prisms, a set of scanning mirrors, a deformable mirror and off-the-shelf optics. The compact double wedge prisms can change the view angles (azimuth angle and zenith angle). A two-axis scanning mirror is integrated into the system to keep the FOV on the target during the change of view angles. The scanning mirror also provides the ability to track a moving object inside the FOV. An improved system with a telecentric lens group is also proposed, which can decouple the rotation angle of scanning mirror and view angles. As an optical system...