基于视觉的穴盘苗茎参数盘上检测装置研究

发布时间：2018-01-18 11:03

本文关键词：基于视觉的穴盘苗茎参数盘上检测装置研究　出处：《浙江理工大学》2017年硕士论文　论文类型：学位论文

【摘要】：嫁接苗的选择直接关系到自动嫁接机的工作效率和嫁接成功率,通过对苗茎参数的测量可为砧木和接穗的自动匹配嫁接提供数据依据。本文开展了基于计算机视觉的苗茎参数检测方法研究,并研制了一种苗茎参数自动检测装置,为苗茎的生长状况的检测、自动嫁接机的砧木和接穗的自动分选创造了条件。主要研究工作包括以下四个方面:1.嫁接苗检测平台总体设计。分析了平台的功能与结构,整个平台由图像采集子系统、辅助测量子系统、运动控制子系统组成,并对各个子系统开展了部件选型分析研究。2.深度检测方法研究。通过前期对穴盘苗尺寸特性调查和检测工作空间的分析,提出了基于激光线高度和距离的几何关系模型的深度检测方法,结合CCD相机和一字激光器设计了一种深度检测辅助装置。经过200组实验,94.5%的测量结果与手工测量相差小于0.1mm误差允许范围,偏差大于0.1mm为5.5%。3.苗茎参数测量方法研究。基于对采集的苗茎图像的通道分离、阈值分割等预处理,开展了激光线的中心线提取、中心线的二次曲线拟合、拟合曲线与苗茎边缘的交点提取等方法研究,以苗茎中轴线为界将苗茎分为左右两个部分分析左右两部分苗径均值,计算得到苗茎值。经过200组测试实验并与手动测量进行比较,最大测量误差为0.241mm,平均测量误差为0.151mm。4.运动控制子系统研究。主要有机械臂运动学求导和电机运动控制,并采用分段运动控制的方法保证机械臂终端的运动轨迹精度。经过100次实验,从控制板发送指令到机械臂运动平均响应时间为41.5ms。
[Abstract]:The selection of grafted seedlings is directly related to the efficiency of the automatic grafting machine and the success rate of grafting. The measurement of seedling stem parameters can provide data basis for automatic matching grafting of stock and scion. In this paper, the method of seedling stem parameter detection based on computer vision is studied. An automatic measuring device for seedling stem parameters was developed, which was used to detect the growth status of seedling stem. The automatic sorting of stock and scion of automatic grafting machine creates conditions. The main research work includes the following four aspects: 1. The overall design of grafted seedling detection platform. The function and structure of the platform are analyzed. The whole platform consists of image acquisition subsystem, auxiliary measurement subsystem and motion control subsystem. And to each subsystem type selection analysis research. 2. Depth detection method research. Through the previous investigation of the hole seedling size characteristics and the analysis of the test workspace. A depth detection method based on the geometric relationship model of laser line height and distance is proposed. A depth detection auxiliary device is designed by combining CCD camera and a word laser. The difference between the measurement results of 94.5% and manual measurement is less than 0.1 mm error allowable range. The deviation greater than 0.1 mm is 5.5 and 3.The method of measuring seedling stem parameters is studied. Based on the pretreatment of channel separation and threshold segmentation of collected seedling stem image, the centerline extraction of laser line is carried out. The conic fitting of the center line and the extraction of the intersection point between the fitting curve and the edge of the seedling stem were studied. The seedling stem was divided into two parts to analyze the mean seedling diameter of the right and left parts with the central axis of the seedling stem as the boundary. The maximum measurement error was 0.241 mm after 200 test groups and compared with manual measurement. The average measurement error is 0.151mm. 4. The research of motion control subsystem mainly includes kinematics derivation of manipulator and motor motion control. After 100 experiments, the average response time from the control board to the robot arm motion response time is 41.5 ms.The method of piecewise motion control is used to ensure the precision of the motion trajectory of the manipulator terminal, and the average response time from the control board to the robot arm motion is 41.5 ms.
【学位授予单位】：浙江理工大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：S223;TP391.41

【参考文献】