Research On Optimization Technique Of Micromotion Measurement Based On Microscopic Vision

Micro motion measurement technology is a new interdisciplinary technology,which plays an important role in biomedical,chemical production,microelectronics manufac-turing and other fields.As a branch of micro/nano technology,micro vision measure-ment technology with its advantages of simple structure,wide measurement range and low cost firmly occupies a very important role,so it has attracted many scholars' re-search interest.However,the key factor restricting the wide application of micro vision measurement technology is that its measuring efficiency is very low due to the high computational complexity.Therefore,this paper proposes an image block matching method based on the improved PSO algorithm for micro vision measurement,which is driven by greatly improving its efficiency without loss of accuracy,and it is demon-strated using the testing data acquired from micro/nano displacement stage.The main work of this paper includes the following contents:(1)This paper studies the micro vision measurement technology for planar mi-cro motion measurement from the perspectives of optical principle,coordinate system transformation and motion analysis decoupling.And a typical micro vision measure-ment system is proposed from the aspects of structure and composition.(2)An improved PSO algorithm is proposed to optimize the process of image block matching.The improved algorithm uses a hierarchical particle swarm migration strate-gy which investigates an inverted pyramid structure and standard local search algorith-m to reduce the computational complexity and ensure high precision.The proposed algorithm can be divided into two phases:a)the first phase aims to create a custom space in the solution space,and then use the inverted pyramid structure to narrow the custom space from the bottom to the top,and accelerate the migration of its external particles to its internal;b)The standard local search algorithm is applied to search the final custom space to achieve a higher accurate result.(3)A novel image block matching algorithm based on distance graph is designed to evaluate the quality of image block matching.This method can be used to measure all three degrees of freedom(DOF)in a plane(including two orthogonal linear DOF and one rotation DOF in a plane).And the PSO algorithm or the improved one proposed in this paper is applied to the image block matching procedure to achieve the motion estimation based on image.(4)A micro vision experimental verification platform for planar micro motion mea-surement is built,and two problems are verified by the motion measurement exper-iments of three-DOF and two-DOF micro/nano stage:a)the performance of the pro-posed method,the standard PSO algorithm,the method in the previous work and the genetic algorithm are compared in terms of the measuring accuracy,efficiency and sta-bility;b)The performance of the six local search algorithms mentioned in the second phase of the proposed method is compared comprehensively from the measurement accuracy and efficiency.