Research On Key Technologies Of Characterization For Acoustic Potential Well Mechanical Property Based On Machine Vision

With the further development of ultrasonic radiation theory and its extensive application, there is an increasing demand of the characterization of the mechanical property of acoustic potential well. And the ultrasonic force detection technology which based on machine vision, because of its rich information and non-contact detection method, is particularly well suited for the characterization of acoustic potential well mechanical property. However, ultrasonic force tends to be produced and used in liquid, which means, the characterization work has to be done in liquid. Since there are still many problems existing in underwater visual imaging, the recognition and tracking of micro-target is very difficult. In the meantime, the building of acoustic kinematics model in liquid is the key point of this technology, but it is not yet clear that the force situation and characteristics of the micro-scale target at theoretical level, and there is no any effective method to calibrate the ultrasonic radiation force, that is why this dissertation developed the research on the key technologies of the characterization of acoustic potential well mechanical property based on machine vision, which was supported by the Key project of Natural Science Foundation of Zhejiang Province "Research on theory and application of automatic 3D micro-assembly technology based on acoustic manipulation". On the basis of studying the mechanical characteristics of micro scale object in water environment, building the acoustic induced kinematics model of the micro-particle in liquid, estimating the motion parameters of micro-particle based on machine vision, calculating the ultrasonic radiation force acting on it based on the motion parameters, the characterization of the mechanical property of acoustic potential well has been realized. The ultrasonic radiation force detection system based on machine vision was constructed and its experimental results demonstrate the feasibility and validity of this technique. The details were presented as follows:Chapter 1:The significance of research on characterization of the acoustic potential well mechanical property was illustrated comprehensively; the current status and development trend of the theory and technique of ultrasonic radiation force was summarized; the key technique of ultrasonic radiation force detection technology based on machine vision was defined; the analysis on the existing problems of the key technologies were carried out, which indicates the studies of the dissertation.Chapter 2:According to scale effect and the forced characteristics of micro scale object, calculating the ultrasonic radiation force acting on micro scale object in arbitrary sound fields, the relation between the motion of spherical micro-particle; besides, the acoustic induced kinematics model of the spherical micro-particle in liquid was build basing on newton’s second law, which established the theoretical foundation of the dissertation.Chapter 3:On the basis of studying of underwater imaging model of micro machine vision and characteristics of underwater image, an improved underwater image enhancement method was developed, and combined with global threshold and morphologic processing; a fast underwater image preprocessing method was realized. Meanwhile, the object recognition algorithm based on pattern matching and the sub-pixel locating algorithm was used to realize object recognition and locating. And also the motion parameters were estimated based on frame difference method.Chapter 4:Research on characterization of the acoustic potential well mechanical property based on machine vision was conducted. According to the acoustic induced kinematics model of the spherical micro-particle and the estimated motion parameters, the radiation force can be calculated, the characterization of the mechanical property of acoustic potential well basing on phase adjusting of the incident acoustic waves can be done. In the meantime, the technique of visualization of the acoustic potential well mechanical property was developed, the conditions for evaluating the acoustic potential well and basis properties of an ideal acoustic well were proposed.Chapter 5:In order to realize the construction of ultrasonic radiation force detection system based on machine vision, on designing of overall technological scheme, the key modules of the system were designed and developed. Then the experiments of characterization the mechanical property of acoustic potential well was conducted, evaluation of the acoustic potential well mechanical property were realized and the results demonstrate the feasibility and validity of the developed technique and system.Chapter 6:The main work of the dissertation was summarized and the prospect for further research was put forward.