| 基于改进Floodfill算法的田垄视觉导航系统 |
| |
| 引用本文： | 谢仕烁,黄伟锋,朱立学,杨尘宇,张世昂,付根平.基于改进Floodfill算法的田垄视觉导航系统[J].中国农机化学报,2021(3). |
| |
| 作者姓名： | 谢仕烁 黄伟锋 朱立学 杨尘宇 张世昂 付根平 |
| |
| 作者单位： | 仲恺农业工程学院自动化学院;仲恺农业工程学院现代农业工程创新研究院;广东省现代精准农业智能化装备技术研发中心;仲恺农业工程学院机电工程学院 |
| |
| 基金项目： | 广东省科技厅重点领域研发计划(2019B020223003);广东省科技创新战略专项资金(KG200320102);广东省大学生创新创业训练计划(S201911347051)。 |
| |
| 摘 要： | 探索在环境多变的田垄中进行视觉导航的方法,针对传统田垄视觉导航方法计算量大且导航效果较一般的问题,为林果作业机器人自主作业能力提供基础,本文提出一种基于机器视觉的田垄导航方法:使用改进的Floodfill算法分割路径信息,通过十字法进行路况分类,进而采用与路况相对应的算法进行导航计算。使用多张路径图片和模拟环境对算法的分割性能和导航能力进行测试,在试验测试中,道路偏移值保持在6 cm内。试验表明,改进Floodfill算法与分类导航法结合的视觉导航方法具有可行性,可为低算力田垄视觉导航方法的探索提供新的方法和思路。
|
| 关 键 词： | 机器视觉 树莓派 田垄导航 Floodfill算法 分类导航 |
Vision navigation system of farm based on improved Floodfill method |
| |
| Xie Shishuo,Huang Weifeng,Zhu Lixue,Yang Chenyu,Zhang Shiang,Fu Genping.Vision navigation system of farm based on improved Floodfill method[J].Chinese Agricultural Mechanization,2021(3). |
| |
| Authors: | Xie Shishuo Huang Weifeng Zhu Lixue Yang Chenyu Zhang Shiang Fu Genping |
| |
| Institution: | (College of Automation,Zhongkai University of Agriculture and Engineering,Guangzhou,510225,China;Modern Agricultural Engineering Innovation Research Institute,Zhongkai University of Agriculture and Engineering,Guangzhou,510225,China;Guangdong modern precision agriculture intelligent equipment technology research and Development Center,Guangzhou,510225,China;College of Mechanical and Electrical Engineering,Zhongkai University of Agriculture and Engineering,Guangzhou,510225,China) |
| |
| Abstract: | This paper explores the method of vision navigation in the field with changeable environment,in order to solve the problem of large amount of calculation and general navigation effect of traditional field vision navigation method and provides the basis for the autonomous operation ability of agricultural robot,figure out a field navigate method based on machine vision:The improved Floodfill algorithmis used to segment the road information,the Cross algorithm is used to classify the road conditions,and thenthe navigation information is calculated by the corresponding method.In this paper,the segmentation performance and the navigation performance are tested by many path images and the imitated environment.In the experiment,the road offset value is kept within 6 cm.The experimental results show that the visual navigation method based on the combination of the improved Floodfill algorithm and the classified navigation algorithm is feasible,which can provide new method and ideas for the exploring the visual navigation method with low computing power. |
| |
| Keywords: | machine vision Raspberry Pi navigation of farm Floodfill method classified navigation |
| 本文献已被 CNKI 维普 等数据库收录！ |
|
