当前位置: 首页 > 文章 > 采摘机器人移动果实目标跟踪研究——基于云存储和无线传感器网络 农机化研究 2019 (2) 206-210
Position: Home > Articles > Research on Moving Fruit Target Tracking of Picking Robot——Based on Cloud Storage and Wireless Sensor Networks Journal of Agricultural Mechanization Research 2019 (2) 206-210

采摘机器人移动果实目标跟踪研究——基于云存储和无线传感器网络

作  者:
胡彬;王超
单  位:
河南工业职业技术学院
关键词:
采摘机器人;目标跟踪;云存储;无线传感网络;图像采集
摘  要:
提高采摘机器人对运动目标的定位能力是提高机器人采摘精度的重要途径,但对于运动果实目标的跟踪和识别需要实时处理大量的图像数据。为有效处理并利用无线传感器实时采集待采摘果实图像,提出了一种基于Hadoop云平台的图像并行处理方案。为了验证方案的可行性,设计了具有运动图像采集和无线传感网络传输功能的采摘机器人,并搭建了基于云存储并行计算的图像抓取平台,利用无线传感器采集的果实图像资源作为原始数据集,对运动待采摘目标进行了图像识别。实验结果表明:采用该方案可以成功地获取运动果实的位置信息,且采摘机器人成功采摘率较高,对于高精度采摘机器人的设计研究具有重要的意义。
译  名:
Research on Moving Fruit Target Tracking of Picking Robot——Based on Cloud Storage and Wireless Sensor Networks
作  者:
Hu Bin;Wang Chao;Henan Polytechnic Institute;
单  位:
Hu Bin%Wang Chao%Henan Polytechnic Institute
关键词:
picking robot;;target tracking;;cloud storage;;wireless sensor network;;image acquisition
摘  要:
In order to improve the localization ability of picking robot for moving target,it is an important way to improve the accuracy of picking robot for motion tracking and recognition,but the fruit of the target with a large number of image data in real time,effective treatment and the use of wireless sensor acquisition for picking fruit image. It proposed an image Hadoop cloud platform based on parallel processing plan. In order to verify the feasibility of design with image acquisition and wireless sensor network transmission function of the picking robot,and build a cloud storage platform based on parallel computation of image capture,image acquisition using fruit resources in wireless sensor as the original data set,the movement for picking target image recognition. The experimental results show that the position information of the moving fruit can be obtained successfully,and the picking rate of the harvesting robot is higher. The scheme is of great significance for the design and research of the high-precision picking robot.

相似文章

计量
文章访问数: 7
HTML全文浏览量: 0
PDF下载量: 0

所属期刊

推荐期刊