智能机器人水果采摘识别系统设计

TP242; 为了提高水果识别的准确性,减少周围环境对识别的影响,设计了一种用于智能水果采摘机器人的识别系统.采用随机森林来学习优化气味传感器的输出,其目的是更好地识别气味;利用摄像机采集到的原始图像计算出对应的本质图像,来对采集到的图像特别是有阴影的图像进行去阴影分析辨识.最后通过结合气味传感器的“气相”结果与摄像机的“色相”结果来评定采摘动作.实验证明该识别系统为后期果实的采摘提供了比较精确的参考....

Full description

Saved in:
Bibliographic Details
Published in计算机应用研究 Vol. 31; no. 9; pp. 2711 - 2714
Main Author 朱霞 陈仁文 夏桦康 章飘艳
Format Journal Article
LanguageChinese
Published 南京航空航天大学机械结构力学及控制国家重点实验室,南京210016 2014
淮阴工学院电子与电气工程学院,江苏淮安223003%南京航空航天大学机械结构力学及控制国家重点实验室,南京,210016
Subjects
智能机器人
机器学习
目标识别
采摘
intelligent robot
object recognition
采摘
机器学习
目标识别
智能机器人
machine learning
picking
Online AccessGet full text
ISSN1001-3695
DOI10.3969/j.issn.1001-3695.2014.09.035

Cover

More Information
Summary:TP242; 为了提高水果识别的准确性,减少周围环境对识别的影响,设计了一种用于智能水果采摘机器人的识别系统.采用随机森林来学习优化气味传感器的输出,其目的是更好地识别气味;利用摄像机采集到的原始图像计算出对应的本质图像,来对采集到的图像特别是有阴影的图像进行去阴影分析辨识.最后通过结合气味传感器的“气相”结果与摄像机的“色相”结果来评定采摘动作.实验证明该识别系统为后期果实的采摘提供了比较精确的参考.
Bibliography:51-1196/TP
machine learning; intelligent robot ; picking ; object recognition
In order to improve the accuracy rates and lower the impact on fruit recognition in environment, this paper presented a recognition system applied in the intelligent fruit picking robot. To recognize the smell more precisely, it optimized the output of the smell sensors by using the random forests algorithm. As to the looks, it analyzed and recognized by removing the shadows of the corresponding intrinsic image which calculated from the original one, especially those with shadows. The robot could de- cide to pick the fruit or not by the ＂smell＂ it smelled with its nose （smell sensors） and the ＂looks＂ it saw with its eyes （came- ras）. Experimental results show that the recognition system can provide a more precise reference for latterly picking.
ZHU Xia, CHEN Ren-wen , XIA Hua-kang , ZHANG Piao-yan（ 1. State Key Laboratory of Mechanics ＆ Control of Mechanical Structures, Nanjing University of Aeronautics ＆ Astronautics, Nanjing 210016,
ISSN:1001-3695
DOI:10.3969/j.issn.1001-3695.2014.09.035