首页 | 本学科首页   官方微博 | 高级检索  
     检索      

激光传感器在喷雾靶标检测中的研究应用
引用本文:沈跃,夏伟,刘慧,李宁.激光传感器在喷雾靶标检测中的研究应用[J].农业现代化研究,2017,38(2):335-342.
作者姓名:沈跃  夏伟  刘慧  李宁
作者单位:江苏大学,江苏大学,江苏大学,江苏大学
基金项目:国家自然科学基金项目(51505195);江苏省自然科学基金项目(BK20130501);江苏省高校自然科学基金 (13KJB210002)
摘    要:树木尺寸、形状等特征信息的获取,是现代化农业中重要的环节,为农林管理中的变量精密喷雾提供依据。为了实现对树木靶标信息的精确测量,搭建了基于UTM-30LX型激光传感器的室内滑台靶标检测系统,实现对规则物体、仿真树的扫描检测,通过编写的C++程序,实时读取数据信息并存储于控制计算机中。利用MATLAB软件对存储的数据信息进行后期的离线分析,重构物体三维图像,通过最大相对误差、轮廓相似度两个参数,分析激光传感器检测得到靶标目标尺寸、重构目标三维图像的精确度。以长方体柜子、圆柱体泡沫、2棵仿真树为试验对象,按照选取的3个检测距离和4个检测速度分别对4个试验对象进行检测与数据分析。试验结果显示,多次试验下,与人工测量尺寸比较而言,激光传感器测得的长方体柜子长度、宽度和高度尺寸最大相对误差分别为6.76%、6.86%和3.92%,圆柱体泡沫的高度、直径尺寸最大相对误差分别为4.25%和7.33%,仿真树1高度、宽度尺寸最大相对误差分别为4.06%和5.91%,仿真树2高度、宽度尺寸最大相对误差分别为3.24%和4.00%;仿真树1、2的激光扫描重构图像与对应实物轮廓相似度最小值分别为0.928 8和0.932 6。研究表明,UTM-30LX型激光传感器在各种试验条件下对多种试验靶标特征检测均具有较好的精确性和准确性,可以担负变量喷雾检测的任务,具有应用到农业生产中的前景。

关 键 词:现代化农业  变量精密施药  激光传感器  靶标检测系统  三维图像
收稿时间:2016/7/18 0:00:00
修稿时间:2016/10/20 0:00:00

Application of laser scanning sensor for spray targets detection
shen yue,Xia Wei,Liu Hui and Li Ning.Application of laser scanning sensor for spray targets detection[J].Research of Agricultural Modernization,2017,38(2):335-342.
Authors:shen yue  Xia Wei  Liu Hui and Li Ning
Institution:Jiangsu university,Jiangsu University,Jiangsu University,Jiangsu University
Abstract:It is a critical step to achieve characteristics of trees in various sizes and shapes for modernizing agriculture applications. Target detection could provide technical data for variable-rate precision spraying of agriculture and forestry management. In order to measure the spray targets accurately, an indoor target detection sliding table platform with UTM-30LX laser sensor was built to detect regular-shape objects and artificial trees. A C language-based algorithm was developed to scan and save laser object data in real time, and an image reconstruction algorithm designed with MATLAB software was used to reconstruct laser object data slices to 3-D canopy images and measure tree sizes. The measurement accuracy of the 3-dimentioal reconstruction images of the spray targets were analyzed by the maximum relative error and edge similarity score. The system accuracy was tested with scanning objects of one rectangular cabinet, one foam cylinder and two artificial trees under indoor laboratory conditions for scanning distances of 2.0 to 2.8 m and sensor travel speeds of 0.3 to 0.9 m/s. Results show that the maximum relative errors about length, width and height of the rectangular cabinet are 6.76%, 6.86% and 3.92% compared with manual measurements in multiple measurements, the maximum relative errors about height and diameter of the foam cylinder are 4.25% and 7.33%, the maximum relative errors about height and width of two artificial trees are 4.06%, 5.91% and 3.24%, 4.00% respectively; The minimum edge similarity scores of the two artificial trees are 0.928 8 and 0.932 6 respectively by several experiments. Results verify the accuracy and precision of UTM-30LX laser sensor on several environments and prove that UTM-30LX laser sensor can be applied in the area of variable-rate precision spraying for the detection of targets.
Keywords:modernizing agriculture  variable-rate precision spraying  laser scanning sensor  spray targets detection system  3-D reconstruction image
本文献已被 CNKI 等数据库收录!
点击此处可从《农业现代化研究》浏览原始摘要信息
点击此处可从《农业现代化研究》下载免费的PDF全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号