The safe use of nuclear energy is an effective way to ease the energy crisis. Compared withnuclear fission, nuclear fusion is non-radioactive and provides more energy per unit mass. And thereare far more material reserves for nuclear fusion on Earth. For these reasons, research on controlled nuclear fusion is always a focus among energy experts all over the world. Since the 1960s, due to theemergence of laser, the inertial confinement fusion, also known as the pellet fusion, has gradually become one of the main ways to achieve controlled nuclear fusion. Micro adhesive dispensing is one of the very important parts of inertial confinement fusion experiments. Currently, the positioning measurement and control of the micro adhesive dispensing is mainly conducted by manual, which is inefficient, high-skill required and unable to meet the needs of practical application. This article is based on the analysis of the micro adhesive dispensing pellet’s hole and inflatable tube. A machine vision-based positioning and control system of micro adhesive dispensing has been developed. Specifically, the work that has been done and the research achievements are as follows: (1) A positioning and control system based on monocular vision has been achieved. The system consists of image acquisition module, image processing module, positioning module and movement control module. The image acquisition module is responsible for the acquisition of the image. The image processing module mainly achieves the image pre-processing and feature extraction. positioning module founds the position of the targets. And then movement control module takes the appropriate actions. (2) A pellet’s hole and inflatable tube’s positioning measurement has been designed. Firstly, use the Auto-focus algorithm to obtain the pellet hole and inflatable tube’s clear images. Secondly, median filter, adaptive threshold segmentation and morphological changes are applied for image pre-processing method. Finally, the pellet’s hole and inflatable tube’s minimum circumscribed rectangular area are extracted and then use the improved ellipse fitting algorithm and RANSAC-based linear fitting algorithm to locate the pellet’s hole and inflatable tube. (3) The calibration of the micro-vision system has been completed. Complete the calibration of the monocular microscopic vision system to find the conversion relationship between the two-dimensional image space with the three-dimensional space. ...
修改评论